File size: 179,189 Bytes
2491c38 |
1 |
[{"sentence": "Here , accuracy is measured by error rate , which is defined as :", "entities": [{"name": "accuracy", "type": "metrics", "pos": [7, 15]}, {"name": "error rate", "type": "metrics", "pos": [31, 41]}]}, {"sentence": "From this perspective , SVM is closely related to other fundamental classification algorithms such as regularized least-squares logistic regression .", "entities": [{"name": "SVM", "type": "algorithm", "pos": [24, 27]}, {"name": "classification algorithms", "type": "else", "pos": [68, 93]}, {"name": "regularized least-squares", "type": "algorithm", "pos": [102, 127]}, {"name": "logistic regression", "type": "algorithm", "pos": [128, 147]}]}, {"sentence": "Brion James portrays Leon Kowalski , a combat and laborer replicant , and Joanna Cassidy portrays Zhora , an assassin replicant .", "entities": [{"name": "Brion James", "type": "person", "pos": [0, 11]}, {"name": "Leon Kowalski", "type": "person", "pos": [21, 34]}, {"name": "Joanna Cassidy", "type": "person", "pos": [74, 88]}, {"name": "Zhora", "type": "person", "pos": [98, 103]}]}, {"sentence": "The first picture to be scanned , stored , and recreated in digital pixels was displayed on the Standards Eastern Automatic Computer ( SEAC ) at NIST .", "entities": [{"name": "Standards Eastern Automatic Computer", "type": "product", "pos": [96, 132]}, {"name": "SEAC", "type": "product", "pos": [135, 139]}, {"name": "NIST", "type": "organization", "pos": [145, 149]}]}, {"sentence": "Segmenting the text into topics or discourse turns might be useful in some natural processing tasks : it can improve information retrieval or speech recognition significantly ( by indexing / recognizing documents more precisely or by giving the specific part of a document corresponding to the query as a result ) .", "entities": [{"name": "Segmenting the text into topics", "type": "task", "pos": [0, 31]}, {"name": "discourse turns", "type": "else", "pos": [35, 50]}, {"name": "information retrieval", "type": "task", "pos": [117, 138]}, {"name": "speech recognition", "type": "task", "pos": [142, 160]}]}, {"sentence": "At Indiana University in 1999 he organized such a symposium , and in April 2000 , he organized a larger symposium entitled Spiritual Robots at Stanford University , in which he moderated a panel consisting of Ray Kurzweil , Hans Moravec , Kevin Kelly , Ralph Merkle , Bill Joy , Frank Drake , John Henry Holland and John Koza .", "entities": [{"name": "Indiana University", "type": "university", "pos": [3, 21]}, {"name": "Spiritual Robots", "type": "conference", "pos": [123, 139]}, {"name": "Stanford University", "type": "university", "pos": [143, 162]}, {"name": "Ray Kurzweil", "type": "researcher", "pos": [209, 221]}, {"name": "Hans Moravec", "type": "researcher", "pos": [224, 236]}, {"name": "Kevin Kelly", "type": "researcher", "pos": [239, 250]}, {"name": "Ralph Merkle", "type": "researcher", "pos": [253, 265]}, {"name": "Bill Joy", "type": "researcher", "pos": [268, 276]}, {"name": "Frank Drake", "type": "researcher", "pos": [279, 290]}, {"name": "John Henry Holland", "type": "researcher", "pos": [293, 311]}, {"name": "John Koza", "type": "researcher", "pos": [316, 325]}]}, {"sentence": "It considers both the precision p and the recall r of the test to compute the score : p is the number of correct positive results divided by the number of all positive results returned by the classifier , and r is the number of correct positive results divided by the number of all relevant samples ( all samples that should have been identified as positive ) .", "entities": [{"name": "precision", "type": "metrics", "pos": [22, 31]}, {"name": "p", "type": "metrics", "pos": [32, 33]}, {"name": "recall", "type": "metrics", "pos": [42, 48]}, {"name": "r", "type": "metrics", "pos": [49, 50]}, {"name": "p", "type": "metrics", "pos": [69, 70]}, {"name": "r", "type": "metrics", "pos": [81, 82]}]}, {"sentence": "Since the Google acquisition , the company has notched up a number of significant achievements , perhaps the most notable being the creation of AlphaGo , a program that defeated world champion Lee Sedol at the complex game of Go .", "entities": [{"name": "Google", "type": "organization", "pos": [10, 16]}, {"name": "AlphaGo", "type": "product", "pos": [144, 151]}, {"name": "Lee Sedol", "type": "person", "pos": [193, 202]}, {"name": "Go", "type": "else", "pos": [226, 228]}]}, {"sentence": "Representing words considering their context through fixed size dense vectors ( word embedding s ) has become one the most fundamental blocks in several NLP systems. an unsupervised disambiguation system uses the similarity between word senses in a fixed context window to select the most suitable word sense using a pre-trained word embedding model and WordNet .", "entities": [{"name": "word embedding", "type": "else", "pos": [80, 94]}, {"name": "NLP", "type": "field", "pos": [153, 156]}, {"name": "unsupervised disambiguation system", "type": "product", "pos": [169, 203]}, {"name": "word sense", "type": "else", "pos": [232, 242]}, {"name": "word embedding", "type": "else", "pos": [329, 343]}, {"name": "WordNet", "type": "product", "pos": [354, 361]}]}, {"sentence": "Machine learning techniques , either Supervised learning or Unsupervised learning , have been used to induce such rules automatically .", "entities": [{"name": "Machine learning", "type": "field", "pos": [0, 16]}, {"name": "Supervised learning", "type": "field", "pos": [37, 56]}, {"name": "Unsupervised learning", "type": "field", "pos": [60, 81]}]}, {"sentence": "In 1969 , Scheinman invented the Stanford arm ,", "entities": [{"name": "Scheinman", "type": "researcher", "pos": [10, 19]}, {"name": "Stanford arm", "type": "product", "pos": [33, 45]}]}, {"sentence": "Since the Log loss is differentiable , a gradient-based method can be used to optimize the model .", "entities": [{"name": "Log loss", "type": "metrics", "pos": [10, 18]}, {"name": "gradient-based method", "type": "else", "pos": [41, 62]}]}, {"sentence": "In machine learning , support-vector machines ( SVMs , also support-vector networks ) are supervised learning models with learning algorithm s that analyze data used for classification and regression analysis .", "entities": [{"name": "machine learning", "type": "field", "pos": [3, 19]}, {"name": "support-vector machines", "type": "algorithm", "pos": [22, 45]}, {"name": "SVMs", "type": "algorithm", "pos": [48, 52]}, {"name": "support-vector networks", "type": "algorithm", "pos": [60, 83]}, {"name": "supervised learning", "type": "field", "pos": [90, 109]}, {"name": "classification", "type": "task", "pos": [170, 184]}, {"name": "regression analysis", "type": "task", "pos": [189, 208]}]}, {"sentence": ", ( 2002 ) as the automatic metric for Machine translation ( MT ) evaluation , many other methods have been proposed to revise or improve it , such as TER , METEOR , Banerjee and Lavie , ( 2005 ) etc .", "entities": [{"name": "Machine translation", "type": "task", "pos": [39, 58]}, {"name": "MT", "type": "task", "pos": [61, 63]}, {"name": "TER", "type": "metrics", "pos": [151, 154]}, {"name": "METEOR", "type": "metrics", "pos": [157, 163]}, {"name": "Banerjee", "type": "researcher", "pos": [166, 174]}, {"name": "Lavie", "type": "researcher", "pos": [179, 184]}]}, {"sentence": "It includes an upper ontology , created by the IEEE working group P1600.1 ( originally by Ian Niles and Adam Pease ) .", "entities": [{"name": "upper ontology", "type": "else", "pos": [15, 29]}, {"name": "IEEE", "type": "organization", "pos": [47, 51]}, {"name": "Ian Niles", "type": "researcher", "pos": [90, 99]}, {"name": "Adam Pease", "type": "researcher", "pos": [104, 114]}]}, {"sentence": "In Cryo Electron Tomography , where the limited number of projections are acquired due to the hardware limitations and to avoid the biological specimen damage , it can be used along with compressive sensing techniques or regularization functions ( e.g. Huber loss ) to improve the reconstruction for better interpretation .", "entities": [{"name": "Cryo Electron Tomography", "type": "else", "pos": [3, 27]}, {"name": "compressive sensing techniques", "type": "algorithm", "pos": [187, 217]}, {"name": "regularization functions", "type": "algorithm", "pos": [221, 245]}, {"name": "Huber loss", "type": "metrics", "pos": [253, 263]}]}, {"sentence": "An implementation of several whitening procedures in R , including ZCA-whitening and PCA whitening but also CCA whitening , is available in the whitening R package published on CRAN .", "entities": [{"name": "whitening", "type": "else", "pos": [29, 38]}, {"name": "R", "type": "program language", "pos": [53, 54]}, {"name": "ZCA-whitening", "type": "algorithm", "pos": [67, 80]}, {"name": "PCA whitening", "type": "algorithm", "pos": [85, 98]}, {"name": "CCA whitening", "type": "algorithm", "pos": [108, 121]}, {"name": "whitening R package", "type": "product", "pos": [144, 163]}, {"name": "CRAN", "type": "product", "pos": [177, 181]}]}, {"sentence": "Today , the field has become even more daunting and complex with the addition of circuit , systems and signal analysis and design languages and software , from MATLAB and Simulink to NumPy , VHDL , PSpice , Verilog and even Assembly language .", "entities": [{"name": "MATLAB", "type": "product", "pos": [160, 166]}, {"name": "Simulink", "type": "product", "pos": [171, 179]}, {"name": "NumPy", "type": "product", "pos": [183, 188]}, {"name": "VHDL", "type": "product", "pos": [191, 195]}, {"name": "PSpice", "type": "product", "pos": [198, 204]}, {"name": "Verilog", "type": "product", "pos": [207, 214]}, {"name": "Assembly language", "type": "program language", "pos": [224, 241]}]}, {"sentence": "The company was founded by Kiichiro Toyoda in 1937 , as a spinoff from Sakichi Toyoda company Toyota Industries to create automobiles .", "entities": [{"name": "Kiichiro Toyoda", "type": "person", "pos": [27, 42]}, {"name": "Sakichi Toyoda", "type": "person", "pos": [71, 85]}, {"name": "Toyota Industries", "type": "organization", "pos": [94, 111]}, {"name": "automobiles", "type": "product", "pos": [122, 133]}]}, {"sentence": "Unsupervised learning , on the other hand , assumes training data that has not been hand-labeled , and attempts to find inherent patterns in the data that can then be used to determine the correct output value for new data instances .. A combination of the two that has recently been explored is semi-supervised learning , which uses a combination of labeled and unlabeled data ( typically a small set of labeled data combined with a large amount of unlabeled data ) .", "entities": [{"name": "Unsupervised learning", "type": "field", "pos": [0, 21]}, {"name": "semi-supervised learning", "type": "field", "pos": [296, 320]}]}, {"sentence": "Despite those humanoid robots for utilitarian uses , there are some humanoid robots which aims at entertainment uses , such as Sony ' s QRIO and Wow Wee ' s RoboSapien .", "entities": [{"name": "Sony", "type": "organization", "pos": [127, 131]}, {"name": "QRIO", "type": "product", "pos": [136, 140]}, {"name": "Wow Wee", "type": "organization", "pos": [145, 152]}, {"name": "RoboSapien", "type": "product", "pos": [157, 167]}]}, {"sentence": "Webber became a Fellow of the Association for the Advancement of Artificial Intelligence in 1991 ,", "entities": [{"name": "Webber", "type": "researcher", "pos": [0, 6]}, {"name": "Association for the Advancement of Artificial Intelligence", "type": "conference", "pos": [30, 88]}]}, {"sentence": "With this company he was developing data-mining and database technology , more specific high-level ontologies for intelligence and automated natural language understanding .", "entities": [{"name": "data-mining", "type": "field", "pos": [36, 47]}, {"name": "database", "type": "field", "pos": [52, 60]}, {"name": "automated natural", "type": "task", "pos": [131, 148]}, {"name": "language understanding", "type": "task", "pos": [149, 171]}]}, {"sentence": "However , in the last years , one can observe appearing of different e-services and related initiatives in developing countries such as Project Nemmadi , MCA21 Mission Mode Project or Digital India even more , in India ; Electronic Government Directorate in Pakistan ; etc .", "entities": [{"name": "Project Nemmadi", "type": "else", "pos": [136, 151]}, {"name": "MCA21 Mission Mode Project", "type": "else", "pos": [154, 180]}, {"name": "Digital India", "type": "else", "pos": [184, 197]}, {"name": "India", "type": "country", "pos": [213, 218]}, {"name": "Electronic Government Directorate", "type": "organization", "pos": [221, 254]}, {"name": "Pakistan", "type": "country", "pos": [258, 266]}]}, {"sentence": "He received a PhD in Radio Physics and Electronics from the Rajabazar Science College campus of University of Calcutta in 1979 as a student of Indian Statistical Institute , and another PhD in Electrical Engineering along with Diploma of the Imperial College from Imperial College , University of London , in 1982 .", "entities": [{"name": "PhD", "type": "else", "pos": [14, 17]}, {"name": "Radio Physics", "type": "field", "pos": [21, 34]}, {"name": "Electronics", "type": "field", "pos": [39, 50]}, {"name": "Rajabazar Science College", "type": "university", "pos": [60, 85]}, {"name": "University of Calcutta", "type": "university", "pos": [96, 118]}, {"name": "Indian Statistical Institute", "type": "university", "pos": [143, 171]}, {"name": "PhD", "type": "else", "pos": [186, 189]}, {"name": "Electrical Engineering", "type": "field", "pos": [193, 215]}, {"name": "Diploma of the Imperial College", "type": "else", "pos": [227, 258]}, {"name": "Imperial College", "type": "university", "pos": [264, 280]}, {"name": "University of London", "type": "university", "pos": [283, 303]}]}, {"sentence": "Expo II was announced as being the locale for the world premiere of several films never before seen in 3D , including The Diamond Wizard and the Universal short , Hawaiian Nights with Mamie Van Doren and Pinky Lee .", "entities": [{"name": "Expo II", "type": "location", "pos": [0, 7]}, {"name": "The Diamond Wizard", "type": "else", "pos": [118, 136]}, {"name": "Hawaiian Nights", "type": "else", "pos": [163, 178]}, {"name": "Mamie Van Doren", "type": "person", "pos": [184, 199]}, {"name": "Pinky Lee", "type": "person", "pos": [204, 213]}]}, {"sentence": "The maximum subarray problem was proposed by Ulf Grenander in 1977 as a simplified model for maximum likelihood estimation of patterns in digitized images .", "entities": [{"name": "Ulf Grenander", "type": "researcher", "pos": [45, 58]}, {"name": "maximum likelihood estimation", "type": "metrics", "pos": [93, 122]}]}, {"sentence": "The iPhone 4S , iPad 3 , iPad Mini 1G , iPad Air , iPad Pro 1G , iPod Touch 5G and later , all come with a more advanced voice assistant called Siri .", "entities": [{"name": "iPhone 4S", "type": "product", "pos": [4, 13]}, {"name": "iPad 3", "type": "product", "pos": [16, 22]}, {"name": "iPad Mini 1G", "type": "product", "pos": [25, 37]}, {"name": "iPad Air", "type": "product", "pos": [40, 48]}, {"name": "iPad Pro 1G", "type": "product", "pos": [51, 62]}, {"name": "iPod Touch 5G", "type": "product", "pos": [65, 78]}, {"name": "Siri", "type": "product", "pos": [144, 148]}]}, {"sentence": "It 's easy to check that the logistic loss and binary cross entropy loss ( Log loss ) are in fact the same ( up to a multiplicative constant math \\ frac { 1 } { \\ log ( 2 ) } / math ) .The cross entropy loss is closely related to the Kullback-Leibler divergence between the empirical distribution and the predicted distribution .", "entities": [{"name": "logistic loss", "type": "metrics", "pos": [29, 42]}, {"name": "binary cross entropy loss", "type": "metrics", "pos": [47, 72]}, {"name": "Log loss", "type": "metrics", "pos": [75, 83]}, {"name": "cross entropy loss", "type": "metrics", "pos": [189, 207]}, {"name": "Kullback-Leibler divergence", "type": "metrics", "pos": [234, 261]}]}, {"sentence": "The EM algorithm is used to find ( local ) maximum likelihood parameters of a statistical model in cases where the equations cannot be solved directly .", "entities": [{"name": "EM algorithm", "type": "algorithm", "pos": [4, 16]}, {"name": "maximum likelihood", "type": "metrics", "pos": [43, 61]}]}, {"sentence": "This research was fundamental to the development of modern techniques of speech synthesis , reading machines for the blind , the study of speech perception and speech recognition , and the development of the motor theory of speech perception .", "entities": [{"name": "speech synthesis", "type": "task", "pos": [73, 89]}, {"name": "reading machines for the blind", "type": "task", "pos": [92, 122]}, {"name": "speech perception", "type": "task", "pos": [138, 155]}, {"name": "speech recognition", "type": "task", "pos": [160, 178]}, {"name": "motor theory of speech perception", "type": "task", "pos": [208, 241]}]}, {"sentence": "The Arduino integrated development environment ( IDE ) is a cross-platform application ( for Windows , macOS , and Linux ) that is written in the programming language Java .", "entities": [{"name": "Arduino", "type": "product", "pos": [4, 11]}, {"name": "integrated development environment", "type": "else", "pos": [12, 46]}, {"name": "IDE", "type": "else", "pos": [49, 52]}, {"name": "cross-platform application", "type": "else", "pos": [60, 86]}, {"name": "Windows", "type": "product", "pos": [93, 100]}, {"name": "macOS", "type": "product", "pos": [103, 108]}, {"name": "Linux", "type": "product", "pos": [115, 120]}, {"name": "Java", "type": "program language", "pos": [167, 171]}]}, {"sentence": "Neural network research stagnated after the publication of machine learning research by Marvin Minsky and Seymour Papert ( 1969 ) .", "entities": [{"name": "Neural network", "type": "algorithm", "pos": [0, 14]}, {"name": "machine learning", "type": "field", "pos": [59, 75]}, {"name": "Marvin Minsky", "type": "researcher", "pos": [88, 101]}, {"name": "Seymour Papert", "type": "researcher", "pos": [106, 120]}]}, {"sentence": "Only a few non-Japanese companies ultimately managed to survive in this market , the major ones being : Adept Technology , Stäubli , the Sweden - Switzerland company ABB Asea Brown Boveri , the Germany company KUKA Robotics and the Italy company Comau .", "entities": [{"name": "Adept Technology", "type": "organization", "pos": [104, 120]}, {"name": "Stäubli", "type": "organization", "pos": [123, 130]}, {"name": "Sweden", "type": "country", "pos": [137, 143]}, {"name": "Switzerland", "type": "country", "pos": [146, 157]}, {"name": "ABB Asea Brown Boveri", "type": "organization", "pos": [166, 187]}, {"name": "Germany", "type": "country", "pos": [194, 201]}, {"name": "KUKA Robotics", "type": "organization", "pos": [210, 223]}, {"name": "Italy", "type": "country", "pos": [232, 237]}, {"name": "Comau", "type": "organization", "pos": [246, 251]}]}, {"sentence": "The research activities include an annual research conference , the RuleML Symposium , also known as RuleML for short .", "entities": [{"name": "RuleML Symposium", "type": "conference", "pos": [68, 84]}, {"name": "RuleML", "type": "conference", "pos": [101, 107]}]}, {"sentence": "Concepts are used as formal tools or models in mathematics , computer science , databases and artificial intelligence where they are sometimes called classes , schema or categories .", "entities": [{"name": "mathematics", "type": "field", "pos": [47, 58]}, {"name": "computer science", "type": "field", "pos": [61, 77]}, {"name": "databases", "type": "field", "pos": [80, 89]}, {"name": "artificial intelligence", "type": "field", "pos": [94, 117]}]}, {"sentence": "He has won awards from the American Psychological Association , the National Academy of Sciences , the Royal , the Cognitive Neuroscience Society and the American Humanist Association .", "entities": [{"name": "American Psychological Association", "type": "organization", "pos": [27, 61]}, {"name": "National Academy of Sciences", "type": "organization", "pos": [68, 96]}, {"name": "Royal", "type": "organization", "pos": [103, 108]}, {"name": "Cognitive Neuroscience Society", "type": "organization", "pos": [115, 145]}, {"name": "American Humanist Association", "type": "organization", "pos": [154, 183]}]}, {"sentence": "Starring Harrison Ford , Rutger Hauer and Sean Young , it is loosely based on Philip K. Dick ' s novel Do Androids Dream of Electric Sheep ? ( 1968 ) .", "entities": [{"name": "Harrison Ford", "type": "person", "pos": [9, 22]}, {"name": "Rutger Hauer", "type": "person", "pos": [25, 37]}, {"name": "Sean Young", "type": "person", "pos": [42, 52]}, {"name": "Philip K. Dick", "type": "person", "pos": [78, 92]}, {"name": "Do Androids Dream of Electric Sheep ?", "type": "else", "pos": [103, 140]}]}, {"sentence": "Image segmentation using k-means clustering algorithms has long been used for pattern recognition , object detection , and medical imaging .", "entities": [{"name": "Image segmentation", "type": "task", "pos": [0, 18]}, {"name": "k-means clustering algorithms", "type": "algorithm", "pos": [25, 54]}, {"name": "pattern recognition", "type": "field", "pos": [78, 97]}, {"name": "object detection", "type": "task", "pos": [100, 116]}, {"name": "medical imaging", "type": "field", "pos": [123, 138]}]}, {"sentence": "General sampling from the truncated normal can be achieved using approximations to the normal CDF and the probit function , and R has a function codertnorm ( ) / code for generating truncated-normal samples .", "entities": [{"name": "CDF", "type": "algorithm", "pos": [94, 97]}, {"name": "probit function", "type": "algorithm", "pos": [106, 121]}, {"name": "R", "type": "program language", "pos": [128, 129]}]}, {"sentence": "He has also received honorary doctorates from the universities of Newcastle , Surrey , Tel Aviv University , , Simon Fraser University and the University of Tromsø .", "entities": [{"name": "universities of Newcastle", "type": "university", "pos": [50, 75]}, {"name": "Surrey", "type": "university", "pos": [78, 84]}, {"name": "Tel Aviv University", "type": "university", "pos": [87, 106]}, {"name": "Simon Fraser University", "type": "university", "pos": [111, 134]}, {"name": "University of Tromsø", "type": "university", "pos": [143, 163]}]}, {"sentence": "A Java implementation using zero based array indexes along with a convenience method for printing the solved order of operations :", "entities": [{"name": "Java", "type": "program language", "pos": [2, 6]}]}, {"sentence": "Such networks are commonly trained under a Cross entropy ( or cross-entropy ) regime , giving a non-linear variant of multinomial logistic regression .", "entities": [{"name": "Cross entropy", "type": "metrics", "pos": [43, 56]}, {"name": "cross-entropy", "type": "metrics", "pos": [62, 75]}, {"name": "multinomial logistic regression", "type": "algorithm", "pos": [118, 149]}]}, {"sentence": "The ACL has a European ( European Chapter of the Association for Computational Linguistics )", "entities": [{"name": "ACL", "type": "conference", "pos": [4, 7]}, {"name": "European", "type": "else", "pos": [14, 22]}, {"name": "European Chapter of the Association for Computational Linguistics", "type": "conference", "pos": [25, 90]}]}, {"sentence": "Two professors , Hal Abelson and Gerald Jay Sussman , chose to remain neutral - their group was referred to variously as Switzerland and Project MAC for the next 30 years .", "entities": [{"name": "Hal Abelson", "type": "researcher", "pos": [17, 28]}, {"name": "Gerald Jay Sussman", "type": "researcher", "pos": [33, 51]}, {"name": "Switzerland", "type": "country", "pos": [121, 132]}, {"name": "Project MAC", "type": "else", "pos": [137, 148]}]}, {"sentence": "Following his PhD , Ghahramani moved to the University of Toronto in 1995 as an ITRC Postdoctoral Fellow in the Artificial Intelligence Lab , working with Geoffrey Hinton .", "entities": [{"name": "PhD", "type": "else", "pos": [14, 17]}, {"name": "Ghahramani", "type": "researcher", "pos": [20, 30]}, {"name": "University of Toronto", "type": "university", "pos": [44, 65]}, {"name": "ITRC", "type": "organization", "pos": [80, 84]}, {"name": "Artificial Intelligence Lab", "type": "organization", "pos": [112, 139]}, {"name": "Geoffrey Hinton", "type": "researcher", "pos": [155, 170]}]}, {"sentence": "Subsequent works focused on addressing these problems , but it was not until the advent of the modern computer and the popularisation of Maximum Likelihood ( MLE ) parameterisation techniques that research really took off .", "entities": [{"name": "Maximum Likelihood", "type": "metrics", "pos": [137, 155]}, {"name": "MLE", "type": "metrics", "pos": [158, 161]}]}, {"sentence": "The series was produced by David Fincher , and starred Kevin Spacey .", "entities": [{"name": "David Fincher", "type": "person", "pos": [27, 40]}, {"name": "Kevin Spacey", "type": "person", "pos": [55, 67]}]}, {"sentence": "Due to limits in computing power , current in silico methods usually must trade speed for accuracy ; e.g. , use rapid protein docking methods instead of computationally costly free energy calculation s .", "entities": [{"name": "accuracy", "type": "metrics", "pos": [90, 98]}, {"name": "protein docking", "type": "algorithm", "pos": [118, 133]}, {"name": "free energy calculation", "type": "algorithm", "pos": [176, 199]}]}, {"sentence": "It had over 30 locations in the U.S. , Canada , Mexico , Brazil and Argentina .", "entities": [{"name": "U.S.", "type": "country", "pos": [32, 36]}, {"name": "Canada", "type": "country", "pos": [39, 45]}, {"name": "Mexico", "type": "country", "pos": [48, 54]}, {"name": "Brazil", "type": "country", "pos": [57, 63]}, {"name": "Argentina", "type": "country", "pos": [68, 77]}]}, {"sentence": "An example of a typical computer vision computation pipeline for Facial recognition system using k -NN including feature extraction and dimension reduction pre-processing steps ( usually implemented with OpenCV ) :", "entities": [{"name": "computer vision", "type": "field", "pos": [24, 39]}, {"name": "Facial recognition system", "type": "product", "pos": [65, 90]}, {"name": "k -NN", "type": "algorithm", "pos": [97, 102]}, {"name": "feature extraction", "type": "task", "pos": [113, 131]}, {"name": "dimension reduction", "type": "task", "pos": [136, 155]}, {"name": "OpenCV", "type": "product", "pos": [204, 210]}]}, {"sentence": "It has a rich set of features , libraries for constraint logic programming , multithreading , unit testing , GUI , interfacing to Java , ODBC and others , literate programming , a web server , SGML , RDF , RDFS , developer tools ( including an IDE with a GUI debugger and GUI profiler ) , and extensive documentation .", "entities": [{"name": "constraint logic programming", "type": "algorithm", "pos": [46, 74]}, {"name": "multithreading", "type": "else", "pos": [77, 91]}, {"name": "unit testing", "type": "else", "pos": [94, 106]}, {"name": "GUI", "type": "else", "pos": [109, 112]}, {"name": "Java", "type": "program language", "pos": [130, 134]}, {"name": "ODBC", "type": "product", "pos": [137, 141]}, {"name": "literate programming", "type": "algorithm", "pos": [155, 175]}, {"name": "web server", "type": "else", "pos": [180, 190]}, {"name": "SGML", "type": "else", "pos": [193, 197]}, {"name": "RDF", "type": "else", "pos": [200, 203]}, {"name": "RDFS", "type": "else", "pos": [206, 210]}, {"name": "IDE", "type": "else", "pos": [244, 247]}, {"name": "GUI debugger", "type": "else", "pos": [255, 267]}, {"name": "GUI profiler", "type": "else", "pos": [272, 284]}]}, {"sentence": "In computer vision and image processing , the notion of scale space representation and Gaussian derivative operators is as a canonical multi-scale representation .", "entities": [{"name": "computer vision", "type": "field", "pos": [3, 18]}, {"name": "image processing", "type": "field", "pos": [23, 39]}, {"name": "scale space representation", "type": "else", "pos": [56, 82]}, {"name": "Gaussian derivative operators", "type": "else", "pos": [87, 116]}, {"name": "canonical multi-scale representation", "type": "else", "pos": [125, 161]}]}, {"sentence": "He is also the President of the Neural Information Processing Systems Foundation , a non-profit organization that oversees the annual Conference on Neural Information Processing Systems Conference .", "entities": [{"name": "Neural Information Processing Systems Foundation", "type": "organization", "pos": [32, 80]}, {"name": "Conference on Neural Information Processing Systems Conference", "type": "conference", "pos": [134, 196]}]}, {"sentence": "For regression analysis problems the squared error can be used as a loss function , for classification the cross entropy can be used .", "entities": [{"name": "regression analysis", "type": "task", "pos": [4, 23]}, {"name": "squared error", "type": "metrics", "pos": [37, 50]}, {"name": "loss function", "type": "else", "pos": [68, 81]}, {"name": "classification", "type": "task", "pos": [88, 102]}, {"name": "cross entropy", "type": "metrics", "pos": [107, 120]}]}, {"sentence": "Lafferty served many prestigious positions , including : 1 ) program co-chair and general co-chair of the Neural Information Processing Systems ( Conference on Neural Information Processing Systems ) Foundation conferences ; 2 ) co-director of CMU 's new Ph.D. Machine Learning Ph.D. Program ; 3 ) associate editor of the Journal of Machine Learning Research", "entities": [{"name": "Lafferty", "type": "researcher", "pos": [0, 8]}, {"name": "Neural Information Processing Systems", "type": "conference", "pos": [106, 143]}, {"name": "Conference on Neural Information Processing Systems", "type": "conference", "pos": [146, 197]}, {"name": "CMU", "type": "university", "pos": [244, 247]}, {"name": "Machine Learning", "type": "field", "pos": [261, 277]}]}, {"sentence": "Convex algorithms , such as AdaBoost and LogitBoost , can be defeated by random noise such they can 't learn basic and learnable combinations of weak hypotheses .", "entities": [{"name": "Convex algorithms", "type": "else", "pos": [0, 17]}, {"name": "AdaBoost", "type": "algorithm", "pos": [28, 36]}, {"name": "LogitBoost", "type": "algorithm", "pos": [41, 51]}]}, {"sentence": "Apertium is a shallow-transfer machine translation system , which uses finite state transducer s for all of its lexical transformations , and hidden Markov model s for part-of-speech tagging or word category disambiguation .", "entities": [{"name": "Apertium", "type": "product", "pos": [0, 8]}, {"name": "shallow-transfer machine translation system", "type": "product", "pos": [14, 57]}, {"name": "finite state transducer", "type": "algorithm", "pos": [71, 94]}, {"name": "hidden Markov model", "type": "algorithm", "pos": [142, 161]}, {"name": "part-of-speech tagging", "type": "task", "pos": [168, 190]}, {"name": "word category disambiguation", "type": "task", "pos": [194, 222]}]}, {"sentence": "The natural gradient of mathE f ( x ) / math , complying with the Fisher information metric ( an informational distance measure between probability distributions and the curvature of the relative entropy ) , now reads", "entities": [{"name": "natural gradient", "type": "else", "pos": [4, 20]}, {"name": "Fisher information metric", "type": "metrics", "pos": [66, 91]}, {"name": "relative entropy", "type": "metrics", "pos": [187, 203]}]}, {"sentence": "The S programming language inspired the systems ' S ' -PLUS and R .", "entities": [{"name": "S programming language", "type": "program language", "pos": [4, 26]}, {"name": "' S ' -PLUS", "type": "product", "pos": [48, 59]}, {"name": "R", "type": "program language", "pos": [64, 65]}]}, {"sentence": "The most influential implementation of Planner was the subset of Planner , called Micro-Planner , implemented by Gerald Jay Sussman , Eugene Charniak and Terry Winograd .", "entities": [{"name": "Planner", "type": "product", "pos": [39, 46]}, {"name": "Planner", "type": "product", "pos": [65, 72]}, {"name": "Micro-Planner", "type": "product", "pos": [82, 95]}, {"name": "Gerald Jay Sussman", "type": "researcher", "pos": [113, 131]}, {"name": "Eugene Charniak", "type": "researcher", "pos": [134, 149]}, {"name": "Terry Winograd", "type": "researcher", "pos": [154, 168]}]}, {"sentence": "In 1779 the Germany - Denmark scientist Christian Gottlieb Kratzenstein won the first prize in a competition announced the Russian Imperial Academy of Sciences and Arts for models he built of the human vocal tract that could produce the five long vowel sounds ( in International Phonetic Alphabet notation :", "entities": [{"name": "Germany", "type": "country", "pos": [12, 19]}, {"name": "Denmark", "type": "country", "pos": [22, 29]}, {"name": "Christian Gottlieb Kratzenstein", "type": "researcher", "pos": [40, 71]}, {"name": "Russian", "type": "else", "pos": [123, 130]}, {"name": "Imperial Academy of Sciences and Arts", "type": "university", "pos": [131, 168]}, {"name": "vocal tract", "type": "else", "pos": [202, 213]}, {"name": "vowel sounds", "type": "else", "pos": [247, 259]}, {"name": "International Phonetic Alphabet", "type": "else", "pos": [265, 296]}]}, {"sentence": "New features in Office XP include smart tags , a selection-based search feature that recognizes different types of text in a document so that users can perform additional actions ; a task pane interface that consolidates popular menu bar commands on the right side of the screen to facilitate quick access to them ; new document collaboration capabilities , support for MSN Groups and SharePoint ; and integrated handwriting recognition and speech recognition capabilities .", "entities": [{"name": "Office XP", "type": "product", "pos": [16, 25]}, {"name": "smart tags", "type": "else", "pos": [34, 44]}, {"name": "selection-based search feature", "type": "else", "pos": [49, 79]}, {"name": "task pane interface", "type": "else", "pos": [183, 202]}, {"name": "document collaboration", "type": "task", "pos": [320, 342]}, {"name": "MSN Groups", "type": "product", "pos": [370, 380]}, {"name": "SharePoint", "type": "product", "pos": [385, 395]}, {"name": "handwriting recognition", "type": "task", "pos": [413, 436]}, {"name": "speech recognition", "type": "task", "pos": [441, 459]}]}, {"sentence": "In many applications the units of these networks apply a sigmoid function as an activation function .", "entities": [{"name": "sigmoid function", "type": "algorithm", "pos": [57, 73]}, {"name": "activation function", "type": "else", "pos": [80, 99]}]}, {"sentence": "In 2001 , Mehler was elected a foreign honorary member of the American Academy of Arts and Sciences , and in 2003 , he was elected a Fellow of the American Association for the Advancement of Science .", "entities": [{"name": "Mehler", "type": "researcher", "pos": [10, 16]}, {"name": "American Academy of Arts and Sciences", "type": "organization", "pos": [62, 99]}, {"name": "American Association for the Advancement of Science", "type": "organization", "pos": [147, 198]}]}, {"sentence": "The extension of this concept to non-binary classifications yields the confusion matrix .", "entities": [{"name": "non-binary classifications", "type": "task", "pos": [33, 59]}, {"name": "confusion matrix", "type": "metrics", "pos": [71, 87]}]}, {"sentence": "An updated measurement noise variance estimate can be obtained from the maximum likelihood calculation", "entities": [{"name": "maximum likelihood", "type": "algorithm", "pos": [72, 90]}]}, {"sentence": "In machine learning , the perceptron is an algorithm for supervised learning of binary classification .", "entities": [{"name": "machine learning", "type": "field", "pos": [3, 19]}, {"name": "perceptron", "type": "algorithm", "pos": [26, 36]}, {"name": "supervised learning", "type": "field", "pos": [57, 76]}, {"name": "binary classification", "type": "task", "pos": [80, 101]}]}, {"sentence": "She has also served as Area Chair of several machine learning and vision conferences including Conference on Neural Information Processing Systems , International Conference on Learning Representations , Conference on Computer Vision and Pattern Recognition , International Conference on Computer Vision , and European Conference on Computer Vision .", "entities": [{"name": "machine learning", "type": "field", "pos": [45, 61]}, {"name": "vision", "type": "field", "pos": [66, 72]}, {"name": "Conference on Neural Information Processing Systems", "type": "conference", "pos": [95, 146]}, {"name": "International Conference on Learning Representations", "type": "conference", "pos": [149, 201]}, {"name": "Conference on Computer Vision and Pattern Recognition", "type": "conference", "pos": [204, 257]}, {"name": "International Conference on Computer Vision", "type": "conference", "pos": [260, 303]}, {"name": "European Conference on Computer Vision", "type": "conference", "pos": [310, 348]}]}, {"sentence": "The condensation algorithm has also been used for facial recognition system in a video sequence .", "entities": [{"name": "condensation algorithm", "type": "algorithm", "pos": [4, 26]}, {"name": "facial recognition system", "type": "product", "pos": [50, 75]}]}, {"sentence": "Information Dissemination is also part of ELRA 's missions which is carried through both the organisation of the conference LREC and the Language Resources and Evaluation Journal edited by Springer .", "entities": [{"name": "Information Dissemination", "type": "task", "pos": [0, 25]}, {"name": "ELRA", "type": "conference", "pos": [42, 46]}, {"name": "LREC", "type": "conference", "pos": [124, 128]}, {"name": "Language Resources and Evaluation Journal", "type": "conference", "pos": [137, 178]}, {"name": "Springer", "type": "conference", "pos": [189, 197]}]}, {"sentence": "In linear time-invariant ( LTI ) system theory , control theory , and in digital signal processing or signal processing , the relationship between the input signal , math \\ displaystyle x ( t ) / math , to output signal , math \\ displaystyle y ( t ) / math , of an LTI system is governed by a convolution operation :", "entities": [{"name": "linear time-invariant ( LTI ) system theory", "type": "field", "pos": [3, 46]}, {"name": "control theory", "type": "field", "pos": [49, 63]}, {"name": "digital signal processing", "type": "field", "pos": [73, 98]}, {"name": "signal processing", "type": "field", "pos": [102, 119]}, {"name": "LTI system", "type": "field", "pos": [265, 275]}, {"name": "convolution", "type": "algorithm", "pos": [293, 304]}]}, {"sentence": "Due to its generality , the field is studied in many other disciplines , such as game theory , control theory , operations research , information theory , simulation-based optimization , multi-agent systems , swarm intelligence , statistics and genetic algorithm s .", "entities": [{"name": "game theory", "type": "field", "pos": [81, 92]}, {"name": "control theory", "type": "field", "pos": [95, 109]}, {"name": "operations research", "type": "field", "pos": [112, 131]}, {"name": "information theory", "type": "field", "pos": [134, 152]}, {"name": "simulation-based optimization", "type": "field", "pos": [155, 184]}, {"name": "multi-agent systems", "type": "product", "pos": [187, 206]}, {"name": "swarm intelligence", "type": "field", "pos": [209, 227]}, {"name": "statistics", "type": "field", "pos": [230, 240]}, {"name": "genetic algorithm", "type": "algorithm", "pos": [245, 262]}]}, {"sentence": "Stochastic gradient descent is a popular algorithm for training a wide range of models in machine learning , including ( linear ) support vector machine s , logistic regression ( see , e.g. , Vowpal Wabbit ) and graphical model s.Jenny Rose Finkel , Alex Kleeman , Christopher D. Manning ( 2008 ) .", "entities": [{"name": "Stochastic gradient descent", "type": "algorithm", "pos": [0, 27]}, {"name": "machine learning", "type": "field", "pos": [90, 106]}, {"name": "support vector machine", "type": "algorithm", "pos": [130, 152]}, {"name": "logistic regression", "type": "algorithm", "pos": [157, 176]}, {"name": "Vowpal Wabbit", "type": "algorithm", "pos": [192, 205]}, {"name": "graphical model", "type": "algorithm", "pos": [212, 227]}, {"name": "s.Jenny Rose Finkel", "type": "researcher", "pos": [228, 247]}, {"name": "Alex Kleeman", "type": "researcher", "pos": [250, 262]}, {"name": "Christopher D. Manning", "type": "researcher", "pos": [265, 287]}]}, {"sentence": "In August 2011 , it was announced that Hitachi would donate an electron microscope to each of five universities in Indonesia ( the University of North Sumatra in Medan , the Indonesian Christian University in Jakarta , Padjadjaran University in Bandung , Jenderal Soedirman University in Purwokerto and Muhammadiyah University in Malang ) .", "entities": [{"name": "Hitachi", "type": "organization", "pos": [39, 46]}, {"name": "electron microscope", "type": "product", "pos": [63, 82]}, {"name": "Indonesia", "type": "country", "pos": [115, 124]}, {"name": "University of North Sumatra", "type": "university", "pos": [131, 158]}, {"name": "Medan", "type": "location", "pos": [162, 167]}, {"name": "Indonesian Christian University", "type": "university", "pos": [174, 205]}, {"name": "Jakarta", "type": "location", "pos": [209, 216]}, {"name": "Padjadjaran University", "type": "university", "pos": [219, 241]}, {"name": "Bandung", "type": "location", "pos": [245, 252]}, {"name": "Jenderal Soedirman University", "type": "university", "pos": [255, 284]}, {"name": "Purwokerto", "type": "location", "pos": [288, 298]}, {"name": "Muhammadiyah University", "type": "university", "pos": [303, 326]}, {"name": "Malang", "type": "location", "pos": [330, 336]}]}, {"sentence": "Optimization techniques of operations research such as linear programming or dynamic programming are often impractical for large scale software engineering problems because of their computational complexity .", "entities": [{"name": "Optimization", "type": "field", "pos": [0, 12]}, {"name": "operations research", "type": "field", "pos": [27, 46]}, {"name": "linear programming", "type": "algorithm", "pos": [55, 73]}, {"name": "dynamic programming", "type": "algorithm", "pos": [77, 96]}, {"name": "software engineering", "type": "field", "pos": [135, 155]}, {"name": "computational complexity", "type": "metrics", "pos": [182, 206]}]}, {"sentence": "Sensitivity is not the same as the precision or positive predictive value ( ratio of TRUE positives to combined TRUE and FALSE positives ) , which is as much a statement about the proportion of actual positives in the population being tested as it is about the test .", "entities": [{"name": "Sensitivity", "type": "metrics", "pos": [0, 11]}, {"name": "precision", "type": "metrics", "pos": [35, 44]}, {"name": "positive predictive value", "type": "metrics", "pos": [48, 73]}, {"name": "TRUE positives", "type": "metrics", "pos": [85, 99]}, {"name": "TRUE and FALSE positives", "type": "metrics", "pos": [112, 136]}]}, {"sentence": "The screenplay by Hampton Fancher ! -- Not titled Android initially - See Sammon , pp. 32 and 38 for explanation -- was optioned in 1977 . Sammon , pp. 23-30 Producer Michael Deeley became interested in Fancher 's draft and convinced director Ridley Scott to film it .", "entities": [{"name": "Hampton Fancher", "type": "person", "pos": [18, 33]}, {"name": "Android", "type": "product", "pos": [50, 57]}, {"name": "Sammon", "type": "person", "pos": [74, 80]}, {"name": "Sammon", "type": "person", "pos": [139, 145]}, {"name": "Michael Deeley", "type": "person", "pos": [167, 181]}, {"name": "Fancher", "type": "person", "pos": [203, 210]}, {"name": "Ridley Scott", "type": "person", "pos": [243, 255]}]}, {"sentence": "Text analysis involves information retrieval , lexical analysis to study word frequency distributions , pattern recognition , tagging / annotation , information extraction , data mining techniques including link and association analysis , visualization , and predictive analytics .", "entities": [{"name": "Text analysis", "type": "field", "pos": [0, 13]}, {"name": "information retrieval", "type": "task", "pos": [23, 44]}, {"name": "lexical analysis", "type": "task", "pos": [47, 63]}, {"name": "word frequency distributions", "type": "else", "pos": [73, 101]}, {"name": "pattern recognition", "type": "field", "pos": [104, 123]}, {"name": "tagging / annotation", "type": "task", "pos": [126, 146]}, {"name": "information extraction", "type": "task", "pos": [149, 171]}, {"name": "data mining", "type": "field", "pos": [174, 185]}, {"name": "link and association analysis", "type": "task", "pos": [207, 236]}, {"name": "visualization", "type": "task", "pos": [239, 252]}, {"name": "predictive analytics", "type": "task", "pos": [259, 279]}]}, {"sentence": "Several metrics use WordNet , a manually constructed lexical database of English words .", "entities": [{"name": "WordNet", "type": "product", "pos": [20, 27]}, {"name": "English", "type": "else", "pos": [73, 80]}]}, {"sentence": "The system uses a combination of techniques from computational linguistics , information retrieval and knowledge representation for finding answers .", "entities": [{"name": "computational linguistics", "type": "field", "pos": [49, 74]}, {"name": "information retrieval", "type": "task", "pos": [77, 98]}, {"name": "knowledge representation for finding answers", "type": "task", "pos": [103, 147]}]}, {"sentence": "As a performance metric , the uncertainty coefficient has the advantage over simple accuracy in that it is not affected by the relative sizes of the different classes .", "entities": [{"name": "uncertainty coefficient", "type": "metrics", "pos": [30, 53]}, {"name": "accuracy", "type": "metrics", "pos": [84, 92]}]}, {"sentence": "Researchers have attempted a number of methods such as optical flow , Kalman filtering , Hidden Markov model s , etc .", "entities": [{"name": "optical flow", "type": "algorithm", "pos": [55, 67]}, {"name": "Kalman filtering", "type": "algorithm", "pos": [70, 86]}, {"name": "Hidden Markov model", "type": "algorithm", "pos": [89, 108]}]}, {"sentence": "She has held the positions of President , Vice President , and Secretary-Treasurer of the Association for Computational Linguistics and has been a board member and secretary of the board of the Computing Research Association .", "entities": [{"name": "Association for Computational Linguistics", "type": "conference", "pos": [90, 131]}, {"name": "Computing Research Association", "type": "organization", "pos": [194, 224]}]}, {"sentence": "Like other similar languages such as APL and MATLAB , R supports matrix arithmetic .", "entities": [{"name": "APL", "type": "program language", "pos": [37, 40]}, {"name": "MATLAB", "type": "product", "pos": [45, 51]}, {"name": "R", "type": "program language", "pos": [54, 55]}, {"name": "matrix arithmetic", "type": "else", "pos": [65, 82]}]}, {"sentence": "On 7 June 2014 , in a Turing test competition at the Royal Society , organised by Kevin Warwick of the University of Reading to mark the 60th anniversary of Turing 's death , Goostman won after 33 % of the judges were convinced that the bot was human .", "entities": [{"name": "Turing test", "type": "else", "pos": [22, 33]}, {"name": "Royal Society", "type": "organization", "pos": [53, 66]}, {"name": "Kevin Warwick", "type": "researcher", "pos": [82, 95]}, {"name": "University of Reading", "type": "university", "pos": [103, 124]}, {"name": "60th anniversary of Turing 's death", "type": "else", "pos": [137, 172]}, {"name": "Goostman", "type": "researcher", "pos": [175, 183]}]}, {"sentence": "A collaborative robot or cobot is a robot that can safely and effectively interact with human workers while performing simple industrial tasks .", "entities": [{"name": "collaborative robot", "type": "product", "pos": [2, 21]}, {"name": "cobot", "type": "product", "pos": [25, 30]}]}, {"sentence": "This overall framework has been applied to a large variety of problems in computer vision , including feature detection , feature classification , image segmentation , image matching , motion estimation , computation of shape cues and object recognition .", "entities": [{"name": "computer vision", "type": "field", "pos": [74, 89]}, {"name": "feature detection", "type": "task", "pos": [102, 119]}, {"name": "feature classification", "type": "task", "pos": [122, 144]}, {"name": "image segmentation", "type": "task", "pos": [147, 165]}, {"name": "image matching", "type": "task", "pos": [168, 182]}, {"name": "motion estimation", "type": "task", "pos": [185, 202]}, {"name": "computation of shape cues", "type": "task", "pos": [205, 230]}, {"name": "object recognition", "type": "task", "pos": [235, 253]}]}, {"sentence": "In many practical applications , parameter estimation for naive Bayes models uses the method of maximum likelihood ; other words , one can work with the naive Bayes model without accepting Bayesian probability or using any Bayesian methods .", "entities": [{"name": "parameter estimation", "type": "task", "pos": [33, 53]}, {"name": "naive Bayes models", "type": "algorithm", "pos": [58, 76]}, {"name": "maximum likelihood", "type": "algorithm", "pos": [96, 114]}, {"name": "naive Bayes", "type": "algorithm", "pos": [153, 164]}, {"name": "Bayesian probability", "type": "algorithm", "pos": [189, 209]}, {"name": "Bayesian methods", "type": "algorithm", "pos": [223, 239]}]}, {"sentence": "Brothers - Victor Gershevich Katz , American mathematician , professor at the Massachusetts Institute of Technology ; Mikhail Gershevich Katz , Israeli mathematician , graduate of Harvard and Columbia ( Ph.D. , 1984 ) universities , professor at Bar-Ilan University , author of the monograph Systolic Geometry and Topology ( Mathematical Surveys and Monographs , vol .", "entities": [{"name": "Victor Gershevich Katz", "type": "researcher", "pos": [11, 33]}, {"name": "American", "type": "else", "pos": [36, 44]}, {"name": "Massachusetts Institute of Technology", "type": "university", "pos": [78, 115]}, {"name": "Mikhail Gershevich Katz", "type": "researcher", "pos": [118, 141]}, {"name": "Israeli", "type": "else", "pos": [144, 151]}, {"name": "Harvard", "type": "university", "pos": [180, 187]}, {"name": "Columbia", "type": "university", "pos": [192, 200]}, {"name": "Ph.D.", "type": "else", "pos": [203, 208]}, {"name": "Bar-Ilan University", "type": "university", "pos": [246, 265]}, {"name": "Systolic Geometry and Topology", "type": "else", "pos": [292, 322]}, {"name": "Mathematical Surveys and Monographs", "type": "else", "pos": [325, 360]}]}, {"sentence": "In 2000 Manuel Toharia , a speaker at previous Campus Parties , and director of Príncipe Felipe 's Museum of Sciences in Valencia 's City of arts and Sciences suggested that Ragageles expand and make the event more international by moving it to the famous museum .", "entities": [{"name": "Manuel Toharia", "type": "person", "pos": [8, 22]}, {"name": "Campus Parties", "type": "conference", "pos": [47, 61]}, {"name": "Príncipe Felipe 's Museum of Sciences", "type": "organization", "pos": [80, 117]}, {"name": "Valencia 's City of arts and Sciences", "type": "location", "pos": [121, 158]}, {"name": "Ragageles", "type": "person", "pos": [174, 183]}]}, {"sentence": "Within 20 minutes , a facial recognition system identifies personal information including family name , ID number and address which are displayed in the street on an advertising screen .", "entities": [{"name": "facial recognition system", "type": "product", "pos": [22, 47]}]}, {"sentence": "Recent research has increasingly focused on unsupervised learning and semi-supervised learning algorithms .", "entities": [{"name": "unsupervised learning", "type": "field", "pos": [44, 65]}, {"name": "semi-supervised learning", "type": "field", "pos": [70, 94]}]}, {"sentence": "Computation of this example using Python code :", "entities": [{"name": "Python", "type": "program language", "pos": [34, 40]}]}, {"sentence": "Today , however , many aspects of speech recognition have been taken over by a deep learning method called Long short-term memory ( LSTM ) , a recurrent neural network published by Sepp Hochreiter & Jürgen Schmidhuber in 1997 .", "entities": [{"name": "speech recognition", "type": "task", "pos": [34, 52]}, {"name": "deep learning", "type": "field", "pos": [79, 92]}, {"name": "Long short-term memory", "type": "algorithm", "pos": [107, 129]}, {"name": "LSTM", "type": "algorithm", "pos": [132, 136]}, {"name": "recurrent neural network", "type": "algorithm", "pos": [143, 167]}, {"name": "Sepp Hochreiter", "type": "researcher", "pos": [181, 196]}, {"name": "Jürgen Schmidhuber", "type": "researcher", "pos": [199, 217]}]}, {"sentence": "In preliminary experimental results with noisy datasets , BrownBoost outperformed AdaBoost ' s generalization error ; however , LogitBoost performed as well as BrownBoost .", "entities": [{"name": "BrownBoost", "type": "algorithm", "pos": [58, 68]}, {"name": "AdaBoost", "type": "algorithm", "pos": [82, 90]}, {"name": "LogitBoost", "type": "algorithm", "pos": [128, 138]}, {"name": "BrownBoost", "type": "algorithm", "pos": [160, 170]}]}, {"sentence": "Evolutionary programming was introduced by Lawrence J. Fogel in the US , while John Henry Holland called his method a genetic algorithm .", "entities": [{"name": "Evolutionary programming", "type": "algorithm", "pos": [0, 24]}, {"name": "Lawrence J. Fogel", "type": "researcher", "pos": [43, 60]}, {"name": "US", "type": "country", "pos": [68, 70]}, {"name": "John Henry Holland", "type": "researcher", "pos": [79, 97]}, {"name": "genetic algorithm", "type": "algorithm", "pos": [118, 135]}]}, {"sentence": "The back-of-the-envelope calculations by Doug , Alan , and their colleagues ( including Marvin Minsky , Allen Newell , Edward Feigenbaum , and John McCarthy ) indicated that that effort would require between 1000 and 3000 person-years of effort , far beyond the standard academic project model .", "entities": [{"name": "Doug", "type": "researcher", "pos": [41, 45]}, {"name": "Alan", "type": "researcher", "pos": [48, 52]}, {"name": "Marvin Minsky", "type": "researcher", "pos": [88, 101]}, {"name": "Allen Newell", "type": "researcher", "pos": [104, 116]}, {"name": "Edward Feigenbaum", "type": "researcher", "pos": [119, 136]}, {"name": "John McCarthy", "type": "researcher", "pos": [143, 156]}]}, {"sentence": "Common criteria are the Mean Squared Error criterion implemented in MSECriterion and the cross-entropy criterion implemented in NLLCriterion .", "entities": [{"name": "Mean Squared Error", "type": "metrics", "pos": [24, 42]}, {"name": "MSECriterion", "type": "metrics", "pos": [68, 80]}, {"name": "cross-entropy criterion", "type": "metrics", "pos": [89, 112]}, {"name": "NLLCriterion", "type": "metrics", "pos": [128, 140]}]}, {"sentence": "Zurada has served the engineering profession as a long-time volunteer of IEEE : as 2014 IEEE Vice-President-Technical Activities ( TAB Chair ) , as President of IEEE Computational Intelligence Society in 2004-05 and the ADCOM member in 2009-14 , 2016-18 and earlier years .", "entities": [{"name": "Zurada", "type": "researcher", "pos": [0, 6]}, {"name": "IEEE", "type": "organization", "pos": [73, 77]}, {"name": "2014 IEEE Vice-President-Technical Activities", "type": "else", "pos": [83, 128]}, {"name": "IEEE Computational Intelligence Society", "type": "conference", "pos": [161, 200]}, {"name": "ADCOM", "type": "conference", "pos": [220, 225]}]}, {"sentence": "In general , computational linguistics draws upon the involvement of linguists , computer science , experts in artificial intelligence , mathematicians , logicians , philosophers , cognitive scientists , cognitive psychologists , psycholinguists , anthropologists and neuroscientists , among others .", "entities": [{"name": "computational linguistics", "type": "field", "pos": [13, 38]}, {"name": "computer science", "type": "field", "pos": [81, 97]}, {"name": "artificial intelligence", "type": "field", "pos": [111, 134]}]}, {"sentence": "Techniques such as dynamic Markov Networks , Convolutional neural network and Long short-term memory are often employed to exploit the inter-frame correlations .", "entities": [{"name": "dynamic Markov Networks", "type": "algorithm", "pos": [19, 42]}, {"name": "Convolutional neural network", "type": "algorithm", "pos": [45, 73]}, {"name": "Long short-term memory", "type": "algorithm", "pos": [78, 100]}]}, {"sentence": "Unimate was the first industrial robot ,", "entities": [{"name": "Unimate", "type": "product", "pos": [0, 7]}, {"name": "industrial robot", "type": "product", "pos": [22, 38]}]}, {"sentence": "Together with Geoffrey Hinton and Yann LeCun , Bengio won the 2018 Turing Award .", "entities": [{"name": "Geoffrey Hinton", "type": "researcher", "pos": [14, 29]}, {"name": "Yann LeCun", "type": "researcher", "pos": [34, 44]}, {"name": "Bengio", "type": "researcher", "pos": [47, 53]}, {"name": "2018 Turing Award", "type": "else", "pos": [62, 79]}]}, {"sentence": "Additional series were filmed at the UK venue for specific sectors of the global market , including two series of Robot Wars Extreme Warriors with United States competitors for the TNN network ( hosted by Mick Foley with Rebecca Grant serving as pit reporter ) , two of Dutch Robot Wars for distribution in the Netherlands and a single series for Germany .", "entities": [{"name": "UK", "type": "country", "pos": [37, 39]}, {"name": "Robot Wars Extreme Warriors", "type": "else", "pos": [114, 141]}, {"name": "United States", "type": "country", "pos": [147, 160]}, {"name": "TNN network", "type": "organization", "pos": [181, 192]}, {"name": "Mick Foley", "type": "person", "pos": [205, 215]}, {"name": "Rebecca Grant", "type": "person", "pos": [221, 234]}, {"name": "Dutch Robot Wars", "type": "else", "pos": [270, 286]}, {"name": "Netherlands", "type": "country", "pos": [311, 322]}, {"name": "Germany", "type": "country", "pos": [347, 354]}]}, {"sentence": "For many years starting from 1986 , Miller directed the development of WordNet , a large computer-readable electronic reference usable in applications such as search engines .", "entities": [{"name": "Miller", "type": "researcher", "pos": [36, 42]}, {"name": "WordNet", "type": "product", "pos": [71, 78]}, {"name": "search engines", "type": "product", "pos": [159, 173]}]}, {"sentence": "Since 2009 , the recurrent neural network s and deep feedforward neural networks developed in the research group of Jürgen Schmidhuber at the Swiss AI Lab IDSIA have won several international handwriting competitions ..", "entities": [{"name": "recurrent neural network", "type": "algorithm", "pos": [17, 41]}, {"name": "deep feedforward neural networks", "type": "algorithm", "pos": [48, 80]}, {"name": "Jürgen Schmidhuber", "type": "researcher", "pos": [116, 134]}, {"name": "Swiss AI Lab IDSIA", "type": "organization", "pos": [142, 160]}, {"name": "international handwriting competitions", "type": "else", "pos": [178, 216]}]}, {"sentence": "The software is implemented in C + + and it is wrapped for Python .", "entities": [{"name": "C + +", "type": "program language", "pos": [31, 36]}, {"name": "Python", "type": "program language", "pos": [59, 65]}]}, {"sentence": "In 1857 , at the request of the Tokugawa Shogunate , a group of Dutch engineers began work on the Nagasaki Yotetsusho , a modern , Western-style foundry and shipyard near the Dutch settlement of Dejima , at Nagasaki .", "entities": [{"name": "Tokugawa Shogunate", "type": "country", "pos": [32, 50]}, {"name": "Dutch", "type": "else", "pos": [64, 69]}, {"name": "Nagasaki Yotetsusho", "type": "else", "pos": [98, 117]}, {"name": "Dutch", "type": "else", "pos": [175, 180]}, {"name": "Dejima", "type": "else", "pos": [195, 201]}, {"name": "Nagasaki", "type": "else", "pos": [207, 215]}]}, {"sentence": "We make as well as possible precise by measuring the mean squared error between mathy / math and math \\ hat { f } ( x ; D ) / math : we want math ( y - \\ hat { f } ( x ; D ) ) ^ 2 / math to be minimal , both for mathx _ 1 , \\ dots , x _ n / math and for points outside of our sample .", "entities": [{"name": "mean squared error", "type": "metrics", "pos": [53, 71]}]}, {"sentence": "He subsequently extended an invitation for Wydner to attend the annual meeting of the American Translators Association that following October where the Weidner Machine Translation System hailed a hoped-for breakthrough in machine translation .", "entities": [{"name": "Wydner", "type": "researcher", "pos": [43, 49]}, {"name": "American Translators Association", "type": "organization", "pos": [86, 118]}, {"name": "Weidner Machine Translation System", "type": "product", "pos": [152, 186]}, {"name": "machine translation", "type": "task", "pos": [222, 241]}]}, {"sentence": "At the 2018 Conference on Neural Information Processing Systems ( NeurIPS ) researchers from Google presented the work .", "entities": [{"name": "2018 Conference on Neural Information Processing Systems", "type": "conference", "pos": [7, 63]}, {"name": "NeurIPS", "type": "conference", "pos": [66, 73]}, {"name": "Google", "type": "organization", "pos": [93, 99]}]}, {"sentence": "The Baum-Welch algorithm uses the well known EM algorithm to find the maximum likelihood estimate of the parameters of a hidden Markov model given a set of observed feature vectors .", "entities": [{"name": "Baum-Welch algorithm", "type": "algorithm", "pos": [4, 24]}, {"name": "EM algorithm", "type": "algorithm", "pos": [45, 57]}, {"name": "maximum likelihood estimate", "type": "metrics", "pos": [70, 97]}, {"name": "hidden Markov model", "type": "algorithm", "pos": [121, 140]}]}, {"sentence": ") In addition to the taxonomic information contained in OpenCyc , ResearchCyc includes significantly more semantic knowledge ( i.e. , additional facts and rules of thumb ) involving the concepts in its knowledge base ; it also includes a large lexicon , English parsing and generation tools , and Java based interfaces for knowledge editing and querying .", "entities": [{"name": "OpenCyc", "type": "product", "pos": [56, 63]}, {"name": "ResearchCyc", "type": "product", "pos": [66, 77]}, {"name": "knowledge base", "type": "else", "pos": [202, 216]}, {"name": "large lexicon , English parsing and generation tools", "type": "product", "pos": [238, 290]}, {"name": "Java", "type": "program language", "pos": [297, 301]}, {"name": "interfaces for knowledge editing and querying", "type": "product", "pos": [308, 353]}]}, {"sentence": "The Hough transform is a feature extraction technique used in image analysis , computer vision , and digital image processing .", "entities": [{"name": "Hough transform", "type": "algorithm", "pos": [4, 19]}, {"name": "feature extraction", "type": "task", "pos": [25, 43]}, {"name": "image analysis", "type": "field", "pos": [62, 76]}, {"name": "computer vision", "type": "field", "pos": [79, 94]}, {"name": "digital image processing", "type": "field", "pos": [101, 125]}]}, {"sentence": "In 1978 , the PUMA ( Programmable Universal Machine for Assembly ) robot was developed by Unimation from Vicarm ( Victor Scheinman ) and with support from General Motors .", "entities": [{"name": "PUMA", "type": "product", "pos": [14, 18]}, {"name": "Programmable Universal Machine for Assembly", "type": "product", "pos": [21, 64]}, {"name": "Unimation", "type": "organization", "pos": [90, 99]}, {"name": "Vicarm", "type": "organization", "pos": [105, 111]}, {"name": "Victor Scheinman", "type": "researcher", "pos": [114, 130]}, {"name": "General Motors", "type": "organization", "pos": [155, 169]}]}, {"sentence": "LSTM was proposed in 1997 by Sepp Hochreiter and Jürgen Schmidhuber .", "entities": [{"name": "LSTM", "type": "algorithm", "pos": [0, 4]}, {"name": "Sepp Hochreiter", "type": "researcher", "pos": [29, 44]}, {"name": "Jürgen Schmidhuber", "type": "researcher", "pos": [49, 67]}]}, {"sentence": "The four outcomes can be formulated in a 2 × 2 contingency table or confusion matrix , as follows :", "entities": [{"name": "contingency table", "type": "metrics", "pos": [47, 64]}, {"name": "confusion matrix", "type": "metrics", "pos": [68, 84]}]}, {"sentence": "He also contributed much through the establishment of ELRA and the LREC conference .", "entities": [{"name": "ELRA", "type": "conference", "pos": [54, 58]}, {"name": "LREC conference", "type": "conference", "pos": [67, 82]}]}, {"sentence": "A popular application for serial robots in today 's industry is the pick-and-place assembly robot , called a SCARA robot , which has four degrees of freedom .", "entities": [{"name": "assembly", "type": "program language", "pos": [83, 91]}, {"name": "SCARA robot", "type": "product", "pos": [109, 120]}]}, {"sentence": "He was one of the founding members and former chair ( 2006-2008 ) of the Special Interest Group on Web as Corpus ( SIGWAC ) of the Association for Computational Linguistics and also one of the founding organizers of SENSEVAL .", "entities": [{"name": "Special Interest Group on Web as Corpus", "type": "conference", "pos": [73, 112]}, {"name": "SIGWAC", "type": "conference", "pos": [115, 121]}, {"name": "Association for Computational Linguistics", "type": "conference", "pos": [131, 172]}, {"name": "SENSEVAL", "type": "conference", "pos": [216, 224]}]}, {"sentence": "As a platform , LinguaStream provides an extensive Java API .", "entities": [{"name": "LinguaStream", "type": "product", "pos": [16, 28]}, {"name": "Java API", "type": "product", "pos": [51, 59]}]}, {"sentence": "The robot kit is Android-based , and it is programmed using Java , the Blocks programming interface , or other Android programming systems .", "entities": [{"name": "Java", "type": "program language", "pos": [60, 64]}, {"name": "Blocks programming interface", "type": "else", "pos": [71, 99]}, {"name": "Android programming systems", "type": "product", "pos": [111, 138]}]}, {"sentence": "The method of defining the linked list specifies the use of a depth-first search or a breadth-first search .", "entities": [{"name": "depth-first search", "type": "algorithm", "pos": [62, 80]}, {"name": "breadth-first search", "type": "algorithm", "pos": [86, 106]}]}, {"sentence": "These regions could signal the presence of objects or parts of objects in the image domain with application to object recognition and / or object video tracking .", "entities": [{"name": "object recognition", "type": "task", "pos": [111, 129]}, {"name": "object video tracking", "type": "task", "pos": [139, 160]}]}, {"sentence": "An example of a semantic network is WordNet , a lexical database of English .", "entities": [{"name": "semantic network", "type": "algorithm", "pos": [16, 32]}, {"name": "WordNet", "type": "product", "pos": [36, 43]}, {"name": "English", "type": "else", "pos": [68, 75]}]}, {"sentence": "Speech recognition is an interdisciplinary subfield of computer science and computational linguistics that develops methodologies and technologies that enable the recognition and translation of spoken language into text by computers .", "entities": [{"name": "Speech recognition", "type": "task", "pos": [0, 18]}, {"name": "computer science", "type": "field", "pos": [55, 71]}, {"name": "computational linguistics", "type": "field", "pos": [76, 101]}, {"name": "recognition and translation of spoken language", "type": "task", "pos": [163, 209]}]}, {"sentence": "Artificial intelligence has retained the most attention regarding applied ontology in subfields like natural language processing within machine and knowledge representation , but ontology editors are being used often in a range of fields like education without the intent to contribute to AI .", "entities": [{"name": "Artificial intelligence", "type": "field", "pos": [0, 23]}, {"name": "applied ontology", "type": "else", "pos": [66, 82]}, {"name": "natural language processing", "type": "field", "pos": [101, 128]}, {"name": "machine", "type": "task", "pos": [136, 143]}, {"name": "knowledge representation", "type": "task", "pos": [148, 172]}, {"name": "AI", "type": "field", "pos": [289, 291]}]}, {"sentence": "This update rule is in fact the stochastic gradient descent update for linear regression .", "entities": [{"name": "stochastic gradient descent", "type": "algorithm", "pos": [32, 59]}, {"name": "linear regression", "type": "algorithm", "pos": [71, 88]}]}, {"sentence": "He was elected to the American Academy of Arts and Sciences and the National Academy of Sciences and has received a series of awards :", "entities": [{"name": "American Academy of Arts and Sciences", "type": "organization", "pos": [22, 59]}, {"name": "National Academy of Sciences", "type": "organization", "pos": [68, 96]}]}, {"sentence": "The most recent school of thought on Honda 's strategy was put forward by Gary Hamel and C. K. Prahalad in 1989 .", "entities": [{"name": "Honda", "type": "organization", "pos": [37, 42]}, {"name": "Gary Hamel", "type": "person", "pos": [74, 84]}, {"name": "C. K. Prahalad", "type": "person", "pos": [89, 103]}]}, {"sentence": "Where BLEU simply calculates n-gram precision adding equal weight to each one , NIST also calculates how informative a particular n-gram is .", "entities": [{"name": "BLEU", "type": "metrics", "pos": [6, 10]}, {"name": "n-gram precision", "type": "metrics", "pos": [29, 45]}, {"name": "NIST", "type": "metrics", "pos": [80, 84]}, {"name": "n-gram", "type": "else", "pos": [130, 136]}]}, {"sentence": "He was honored with the 2019 Lifetime Achievement Award from the Association for Computational Linguistics ( ACL ) .", "entities": [{"name": "2019 Lifetime Achievement Award", "type": "else", "pos": [24, 55]}, {"name": "Association for Computational Linguistics", "type": "conference", "pos": [65, 106]}, {"name": "ACL", "type": "conference", "pos": [109, 112]}]}, {"sentence": "Sycara is a Fellow of Institute of Electrical and Electronics Engineers ( IEEE ) , and a Fellow of American Association for Artificial Intelligence ( AAAI ) .", "entities": [{"name": "Sycara", "type": "researcher", "pos": [0, 6]}, {"name": "Institute of Electrical and Electronics Engineers", "type": "organization", "pos": [22, 71]}, {"name": "IEEE", "type": "organization", "pos": [74, 78]}, {"name": "American Association for Artificial Intelligence", "type": "conference", "pos": [99, 147]}, {"name": "AAAI", "type": "conference", "pos": [150, 154]}]}, {"sentence": "The following MATLAB code demonstrates a concrete solution for solving the non-linear system of equations presented in the previous section : See also", "entities": [{"name": "MATLAB", "type": "product", "pos": [14, 20]}, {"name": "non-linear system", "type": "else", "pos": [75, 92]}]}, {"sentence": "Pattern recognition systems are in many cases trained from labeled training data ( supervised learning ) but when no labeled data are available other algorithms can be used to discover previously unknown patterns ( unsupervised learning ) .", "entities": [{"name": "Pattern recognition systems", "type": "product", "pos": [0, 27]}, {"name": "supervised learning", "type": "field", "pos": [83, 102]}, {"name": "unsupervised learning", "type": "field", "pos": [215, 236]}]}, {"sentence": "It was first used by Lawrence J. Fogel in the US in 1960 in order to use simulated evolution as a learning process aiming to generate artificial intelligence .", "entities": [{"name": "Lawrence J. Fogel", "type": "researcher", "pos": [21, 38]}, {"name": "US", "type": "country", "pos": [46, 48]}, {"name": "artificial intelligence", "type": "field", "pos": [134, 157]}]}, {"sentence": "Reinforcement learning is one of three basic machine learning paradigms , alongside supervised learning and unsupervised learning .", "entities": [{"name": "Reinforcement learning", "type": "field", "pos": [0, 22]}, {"name": "machine learning", "type": "field", "pos": [45, 61]}, {"name": "supervised learning", "type": "field", "pos": [84, 103]}, {"name": "unsupervised learning", "type": "field", "pos": [108, 129]}]}, {"sentence": "In such cases , cloud computing and open source programming language R can help smaller banks to adopt risk analytics and support branch level monitoring by applying predictive analytics .", "entities": [{"name": "cloud computing", "type": "field", "pos": [16, 31]}, {"name": "R", "type": "program language", "pos": [69, 70]}, {"name": "predictive analytics", "type": "field", "pos": [166, 186]}]}, {"sentence": "One of the first versions of the theorem was proved by George Cybenko in 1989 for sigmoid function activation functions. Cybenko G. ( 1989 ) , 2 ( 4 ) , 303-314 .", "entities": [{"name": "George Cybenko", "type": "researcher", "pos": [55, 69]}, {"name": "sigmoid function", "type": "algorithm", "pos": [82, 98]}, {"name": "Cybenko G.", "type": "researcher", "pos": [121, 131]}]}, {"sentence": "In this process , which is known as cross-validation , the MSE is often called the mean squared prediction error , and is computed as", "entities": [{"name": "cross-validation", "type": "algorithm", "pos": [36, 52]}, {"name": "MSE", "type": "metrics", "pos": [59, 62]}, {"name": "mean squared prediction error", "type": "metrics", "pos": [83, 112]}]}, {"sentence": "OMR is generally distinguished from optical character recognition ( OCR ) by the fact that a complicated pattern recognition engine is not required .", "entities": [{"name": "OMR", "type": "task", "pos": [0, 3]}, {"name": "optical character recognition", "type": "task", "pos": [36, 65]}, {"name": "OCR", "type": "task", "pos": [68, 71]}, {"name": "pattern recognition", "type": "field", "pos": [105, 124]}]}, {"sentence": "In 2018 and 2019 , the Championship was be held in Houston and Detroit , Michigan at the TCF Center and Ford Field .", "entities": [{"name": "Houston", "type": "location", "pos": [51, 58]}, {"name": "Detroit", "type": "location", "pos": [63, 70]}, {"name": "Michigan", "type": "location", "pos": [73, 81]}, {"name": "TCF Center", "type": "location", "pos": [89, 99]}, {"name": "Ford Field", "type": "location", "pos": [104, 114]}]}, {"sentence": "Classification can be thought of as two separate problems - binary classification and multiclass classification .", "entities": [{"name": "Classification", "type": "task", "pos": [0, 14]}, {"name": "binary classification", "type": "task", "pos": [60, 81]}, {"name": "multiclass classification", "type": "task", "pos": [86, 111]}]}, {"sentence": "Two examples of popular parallel robots are the Stewart platform and the Delta robot .", "entities": [{"name": "parallel robots", "type": "product", "pos": [24, 39]}, {"name": "Stewart platform", "type": "product", "pos": [48, 64]}, {"name": "Delta robot", "type": "product", "pos": [73, 84]}]}, {"sentence": "( Nevertheless , the ReLU activation function , which is non-differentiable at 0 , has become quite popular , e.g. in AlexNet )", "entities": [{"name": "ReLU activation function", "type": "algorithm", "pos": [21, 45]}, {"name": "AlexNet", "type": "algorithm", "pos": [118, 125]}]}, {"sentence": "The F-score is often used in the field of information retrieval for measuring search , document classification , and query classification performance. and so F_beta is seen in wide application .", "entities": [{"name": "F-score", "type": "metrics", "pos": [4, 11]}, {"name": "information retrieval", "type": "task", "pos": [42, 63]}, {"name": "search", "type": "task", "pos": [78, 84]}, {"name": "document classification", "type": "task", "pos": [87, 110]}, {"name": "query classification", "type": "task", "pos": [117, 137]}, {"name": "F_beta", "type": "metrics", "pos": [158, 164]}]}, {"sentence": "This is done by modeling the received signal then using a statistical estimation method such as maximum likelihood ( ML ) , majority voting ( MV ) or maximum a posteriori ( MAP ) to make a decision about which target in the library best fits the model built using the received signal .", "entities": [{"name": "maximum likelihood", "type": "algorithm", "pos": [96, 114]}, {"name": "ML", "type": "algorithm", "pos": [117, 119]}, {"name": "majority voting", "type": "algorithm", "pos": [124, 139]}, {"name": "MV", "type": "algorithm", "pos": [142, 144]}, {"name": "maximum a posteriori", "type": "algorithm", "pos": [150, 170]}, {"name": "MAP", "type": "algorithm", "pos": [173, 176]}]}, {"sentence": "Sowa received a BS in mathematics from Massachusetts Institute of Technology in 1962 , an MA in applied from Harvard University in 1966 , and a PhD in computer science from the Vrije Universiteit Brussel in 1999 on a dissertation titled Knowledge Representation : Logical , Philosophical , and Computational Foundations .", "entities": [{"name": "Sowa", "type": "researcher", "pos": [0, 4]}, {"name": "BS", "type": "else", "pos": [16, 18]}, {"name": "mathematics", "type": "field", "pos": [22, 33]}, {"name": "Massachusetts Institute of Technology", "type": "university", "pos": [39, 76]}, {"name": "MA", "type": "else", "pos": [90, 92]}, {"name": "applied", "type": "field", "pos": [96, 103]}, {"name": "Harvard University", "type": "university", "pos": [109, 127]}, {"name": "PhD", "type": "else", "pos": [144, 147]}, {"name": "computer science", "type": "field", "pos": [151, 167]}, {"name": "Vrije Universiteit Brussel", "type": "university", "pos": [177, 203]}, {"name": "Knowledge Representation : Logical , Philosophical , and Computational Foundations", "type": "else", "pos": [237, 319]}]}, {"sentence": "Since paraphrase recognition can be posed as a classification problem , most standard evaluations metrics such as accuracy , f1 score , or an ROC curve do relatively well .", "entities": [{"name": "paraphrase recognition", "type": "task", "pos": [6, 28]}, {"name": "classification", "type": "task", "pos": [47, 61]}, {"name": "accuracy", "type": "metrics", "pos": [114, 122]}, {"name": "f1 score", "type": "metrics", "pos": [125, 133]}, {"name": "ROC curve", "type": "metrics", "pos": [142, 151]}]}, {"sentence": "This makes it practical for analyzing large data sets ( hundreds or thousands of taxa ) and for bootstrapping , for which purposes other means of analysis ( e.g. maximum parsimony , maximum likelihood ) may be computation ally prohibitive .", "entities": [{"name": "bootstrapping", "type": "algorithm", "pos": [96, 109]}, {"name": "maximum parsimony", "type": "algorithm", "pos": [162, 179]}, {"name": "maximum likelihood", "type": "algorithm", "pos": [182, 200]}]}, {"sentence": "The 2002 submission of the DAML + OIL language to the World Wide Web Consortium ( W3C ) the work done by DAML contractors and the European Union / United States ad hoc Joint Committee on Markup Languages .", "entities": [{"name": "DAML", "type": "program language", "pos": [27, 31]}, {"name": "OIL", "type": "program language", "pos": [34, 37]}, {"name": "World Wide Web Consortium", "type": "organization", "pos": [54, 79]}, {"name": "W3C", "type": "organization", "pos": [82, 85]}, {"name": "DAML", "type": "program language", "pos": [105, 109]}, {"name": "European Union / United States ad hoc Joint Committee on Markup Languages", "type": "organization", "pos": [130, 203]}]}, {"sentence": "An example of non-linear normalization is when the normalization follows a sigmoid function , in that case , the normalized image is computed according to the formula", "entities": [{"name": "non-linear normalization", "type": "else", "pos": [14, 38]}, {"name": "normalization", "type": "else", "pos": [51, 64]}, {"name": "sigmoid function", "type": "algorithm", "pos": [75, 91]}]}, {"sentence": "It has been pointed out that precision is usually twinned with recall to overcome this problem", "entities": [{"name": "precision", "type": "metrics", "pos": [29, 38]}, {"name": "recall", "type": "metrics", "pos": [63, 69]}]}, {"sentence": "The commonly used metrics are the mean squared error and root mean squared error , the latter having been used in the Netflix Prize .", "entities": [{"name": "mean squared error", "type": "metrics", "pos": [34, 52]}, {"name": "root mean squared error", "type": "metrics", "pos": [57, 80]}, {"name": "Netflix Prize", "type": "else", "pos": [118, 131]}]}, {"sentence": "In August 2016 , a research programme with University College Hospital was announced with the aim of developing an algorithm that can automatically differentiate between healthy and cancerous tissues in head and neck areas .", "entities": [{"name": "University College Hospital", "type": "organization", "pos": [43, 70]}]}, {"sentence": "The impact of Posner 's theoretical and empirical contributions has been recognized through fellowship in the American Psychological Association , the Association for Psychological Science , the Society of Experimental Psychologists , the American Academy of Arts and Sciences , the American Association for the Advancement of Science , and the National Academy of Sciences .", "entities": [{"name": "Posner", "type": "researcher", "pos": [14, 20]}, {"name": "American Psychological Association", "type": "organization", "pos": [110, 144]}, {"name": "Association for Psychological Science", "type": "organization", "pos": [151, 188]}, {"name": "Society of Experimental Psychologists", "type": "organization", "pos": [195, 232]}, {"name": "American Academy of Arts and Sciences", "type": "organization", "pos": [239, 276]}, {"name": "American Association for the Advancement of Science", "type": "organization", "pos": [283, 334]}, {"name": "National Academy of Sciences", "type": "organization", "pos": [345, 373]}]}, {"sentence": "These Intelligent Chatbots make use of all kinds of artificial intelligence like image moderation and natural language understanding ( NLU ) , natural language generation ( NLG ) , machine learning and deep learning .", "entities": [{"name": "Chatbots", "type": "product", "pos": [18, 26]}, {"name": "artificial intelligence", "type": "field", "pos": [52, 75]}, {"name": "image moderation", "type": "task", "pos": [81, 97]}, {"name": "natural language understanding", "type": "task", "pos": [102, 132]}, {"name": "NLU", "type": "task", "pos": [135, 138]}, {"name": "natural language generation", "type": "task", "pos": [143, 170]}, {"name": "NLG", "type": "task", "pos": [173, 176]}, {"name": "machine learning", "type": "field", "pos": [181, 197]}, {"name": "deep learning", "type": "field", "pos": [202, 215]}]}, {"sentence": "The row ratios are Positive Predictive Value ( PPV , aka precision ) ( TP / ( TP + FP ) ) , with complement the FALSE Discovery Rate ( FDR ) ( FP / ( TP + FP ) ) ; and Negative Predictive Value ( NPV ) ( TN / ( TN + FN ) ) , with complement the FALSE Omission Rate ( FOR ) ( FN / ( TN + FN ) ) .", "entities": [{"name": "Positive Predictive Value", "type": "metrics", "pos": [19, 44]}, {"name": "PPV", "type": "metrics", "pos": [47, 50]}, {"name": "precision", "type": "metrics", "pos": [57, 66]}, {"name": "TP / ( TP + FP )", "type": "metrics", "pos": [71, 87]}, {"name": "FALSE Discovery Rate", "type": "metrics", "pos": [112, 132]}, {"name": "FDR", "type": "metrics", "pos": [135, 138]}, {"name": "FP / ( TP + FP )", "type": "metrics", "pos": [143, 159]}, {"name": "Negative Predictive Value", "type": "metrics", "pos": [168, 193]}, {"name": "NPV", "type": "metrics", "pos": [196, 199]}, {"name": "TN / ( TN + FN )", "type": "metrics", "pos": [204, 220]}, {"name": "FALSE Omission Rate", "type": "metrics", "pos": [245, 264]}, {"name": "FOR", "type": "metrics", "pos": [267, 270]}, {"name": "FN / ( TN + FN )", "type": "metrics", "pos": [275, 291]}]}, {"sentence": "The information is a blend of sitemaps and RSS and is created using the Information Model ( IM ) and Biomedical Resource Ontology ( BRO ) .", "entities": [{"name": "RSS", "type": "else", "pos": [43, 46]}, {"name": "Information Model", "type": "algorithm", "pos": [72, 89]}, {"name": "IM", "type": "algorithm", "pos": [92, 94]}, {"name": "Biomedical Resource Ontology", "type": "algorithm", "pos": [101, 129]}, {"name": "BRO", "type": "algorithm", "pos": [132, 135]}]}, {"sentence": "Recent text recognition is based on Recurrent neural network ( Long short-term memory ) and does not require a language model .", "entities": [{"name": "text recognition", "type": "task", "pos": [7, 23]}, {"name": "Recurrent neural network", "type": "algorithm", "pos": [36, 60]}, {"name": "Long short-term memory", "type": "algorithm", "pos": [63, 85]}, {"name": "language model", "type": "algorithm", "pos": [111, 125]}]}, {"sentence": "Popular loss functions include the hinge loss ( for linear SVMs ) and the log loss ( for logistic regression ) .", "entities": [{"name": "loss functions", "type": "else", "pos": [8, 22]}, {"name": "hinge loss", "type": "metrics", "pos": [35, 45]}, {"name": "linear SVMs", "type": "algorithm", "pos": [52, 63]}, {"name": "log loss", "type": "metrics", "pos": [74, 82]}, {"name": "logistic regression", "type": "algorithm", "pos": [89, 108]}]}, {"sentence": "SSIM is designed to improve on traditional methods such as peak signal-to-noise ratio ( PSNR ) and mean squared error ( MSE ) .", "entities": [{"name": "SSIM", "type": "metrics", "pos": [0, 4]}, {"name": "peak signal-to-noise ratio", "type": "metrics", "pos": [59, 85]}, {"name": "PSNR", "type": "metrics", "pos": [88, 92]}, {"name": "mean squared error", "type": "metrics", "pos": [99, 117]}, {"name": "MSE", "type": "metrics", "pos": [120, 123]}]}, {"sentence": "His work inspired subsequent generations of robotics researchers such as Rodney Brooks , Hans Moravec and Mark Tilden .", "entities": [{"name": "Rodney Brooks", "type": "researcher", "pos": [73, 86]}, {"name": "Hans Moravec", "type": "researcher", "pos": [89, 101]}, {"name": "Mark Tilden", "type": "researcher", "pos": [106, 117]}]}, {"sentence": "Further pulse training is not differentiable , eliminating backpropagation -based training methods like gradient descent .", "entities": [{"name": "backpropagation", "type": "algorithm", "pos": [59, 74]}, {"name": "gradient descent", "type": "algorithm", "pos": [104, 120]}]}, {"sentence": "This relations can be easily represented with a confusion matrix , a table which describes the accuracy of a classification model .", "entities": [{"name": "confusion matrix", "type": "metrics", "pos": [48, 64]}, {"name": "accuracy", "type": "metrics", "pos": [95, 103]}, {"name": "classification", "type": "task", "pos": [109, 123]}]}, {"sentence": "At the 2018 Conference on Neural Information Processing Systems ( NeurIPS ) researchers from Google presented the work", "entities": [{"name": "2018 Conference on Neural Information Processing Systems", "type": "conference", "pos": [7, 63]}, {"name": "NeurIPS", "type": "conference", "pos": [66, 73]}, {"name": "Google", "type": "organization", "pos": [93, 99]}]}, {"sentence": "During his time at Duke , he worked on an automated crossword solver PROVERB , which won an Outstanding Paper Award in 1999 from AAAI and competed in the American Crossword Puzzle Tournament .", "entities": [{"name": "Duke", "type": "university", "pos": [19, 23]}, {"name": "PROVERB", "type": "product", "pos": [69, 76]}, {"name": "Outstanding Paper Award", "type": "else", "pos": [92, 115]}, {"name": "AAAI", "type": "conference", "pos": [129, 133]}, {"name": "American Crossword Puzzle Tournament", "type": "else", "pos": [154, 190]}]}, {"sentence": "Headquartered in Rochester Hills , Michigan , the company had 10 regional locations in the U.S. , Canada , Mexico and Brazil .", "entities": [{"name": "Rochester Hills", "type": "location", "pos": [17, 32]}, {"name": "Michigan", "type": "location", "pos": [35, 43]}, {"name": "the U.S.", "type": "country", "pos": [87, 95]}, {"name": "Canada", "type": "country", "pos": [98, 104]}, {"name": "Mexico", "type": "country", "pos": [107, 113]}, {"name": "Brazil", "type": "country", "pos": [118, 124]}]}, {"sentence": "It joins a collection of historically important robots that includes an early Unimate and the Odetics Odex 1 .", "entities": [{"name": "Unimate", "type": "product", "pos": [78, 85]}, {"name": "Odetics Odex 1", "type": "product", "pos": [94, 108]}]}, {"sentence": "A guest editor for that issue will be David 's former colleague at NIST , Judah Levine who is the most recent recipient of the I. I. Rabi Award .", "entities": [{"name": "David", "type": "researcher", "pos": [38, 43]}, {"name": "NIST", "type": "organization", "pos": [67, 71]}, {"name": "Judah Levine", "type": "researcher", "pos": [74, 86]}, {"name": "I. I. Rabi Award", "type": "else", "pos": [127, 143]}]}, {"sentence": "These can be arranged into a 2 × 2 contingency table ( confusion matrix ) , conventionally with the test result on the vertical axis and the actual condition on the horizontal axis .", "entities": [{"name": "confusion matrix", "type": "metrics", "pos": [55, 71]}]}, {"sentence": "The Apple iOS operating system used on the iPhone , iPad and iPod Touch uses VoiceOver speech synthesis accessibility .", "entities": [{"name": "Apple iOS operating system", "type": "product", "pos": [4, 30]}, {"name": "iPhone", "type": "product", "pos": [43, 49]}, {"name": "iPad", "type": "product", "pos": [52, 56]}, {"name": "iPod Touch", "type": "product", "pos": [61, 71]}, {"name": "VoiceOver speech synthesis", "type": "product", "pos": [77, 103]}]}, {"sentence": "For example , the best system entering MUC-7 scored 93.39 % of F-measure while human annotators scored 97.6 % and 96.95 % .", "entities": [{"name": "MUC-7", "type": "conference", "pos": [39, 44]}, {"name": "F-measure", "type": "metrics", "pos": [63, 72]}]}, {"sentence": "This is done using standard neural net training algorithms such as stochastic gradient descent with backpropagation .", "entities": [{"name": "stochastic gradient descent", "type": "algorithm", "pos": [67, 94]}, {"name": "backpropagation", "type": "algorithm", "pos": [100, 115]}]}, {"sentence": "Rotten Tomatoes is a top 1000 site , placing around # 400 globally and top 150 for the US only , according to website ranker Alexa .", "entities": [{"name": "Rotten Tomatoes", "type": "organization", "pos": [0, 15]}, {"name": "US", "type": "country", "pos": [87, 89]}, {"name": "Alexa", "type": "product", "pos": [125, 130]}]}, {"sentence": "Generally speaking all learning displays incremental change over time , but describes an Sigmoid function which has different appearances depending on the time scale of observation .", "entities": [{"name": "Sigmoid function", "type": "algorithm", "pos": [89, 105]}]}, {"sentence": "The SSD is also known as mean squared error .", "entities": [{"name": "SSD", "type": "metrics", "pos": [4, 7]}, {"name": "mean squared error", "type": "metrics", "pos": [25, 43]}]}, {"sentence": "Decision tree learning , neural networks , or a naive Bayes classifier could be used in combination with measures of model quality such as balanced accuracy", "entities": [{"name": "Decision tree learning", "type": "algorithm", "pos": [0, 22]}, {"name": "neural networks", "type": "algorithm", "pos": [25, 40]}, {"name": "naive Bayes classifier", "type": "algorithm", "pos": [48, 70]}]}, {"sentence": "He is a past President ( 1979 ) and an inaugural Fellow ( 2011 ) of the ACL , a co-recipient of the 1992 Association for Computing Machinery Software Systems Award for his contribution to the Interlisp programming system , and a Fellow of the Association for Computing Machinery .", "entities": [{"name": "ACL", "type": "conference", "pos": [72, 75]}, {"name": "1992 Association for Computing Machinery", "type": "conference", "pos": [100, 140]}, {"name": "Software Systems Award", "type": "else", "pos": [141, 163]}, {"name": "Interlisp programming system", "type": "product", "pos": [192, 220]}, {"name": "Association for Computing Machinery", "type": "conference", "pos": [243, 278]}]}, {"sentence": "Along with Geoffrey Hinton and Yann LeCun , Bengio is considered by Cade Metz as one of the three people most responsible for the advancement of deep learning during the 1990s and 2000s .", "entities": [{"name": "Geoffrey Hinton", "type": "researcher", "pos": [11, 26]}, {"name": "Yann LeCun", "type": "researcher", "pos": [31, 41]}, {"name": "Bengio", "type": "researcher", "pos": [44, 50]}, {"name": "Cade Metz", "type": "researcher", "pos": [68, 77]}, {"name": "deep learning", "type": "field", "pos": [145, 158]}]}, {"sentence": "In information theory and computer science , a code is usually considered as an algorithm that uniquely represents symbols from some source alphabet , by encoded strings , which may be in some other target alphabet .", "entities": [{"name": "information theory", "type": "field", "pos": [3, 21]}, {"name": "computer science", "type": "field", "pos": [26, 42]}]}, {"sentence": "A fairly simple non-linear function , the sigmoid function such as the logistic function also has an easily calculated derivative , which can be important when calculating the weight updates in the network .", "entities": [{"name": "sigmoid function", "type": "algorithm", "pos": [42, 58]}, {"name": "logistic function", "type": "algorithm", "pos": [71, 88]}]}, {"sentence": "Čapek was born in Hronov , Bohemia ( Austria-Hungary , later Czechoslovakia , now the Czech Republic ) in 1887 .", "entities": [{"name": "Čapek", "type": "person", "pos": [0, 5]}, {"name": "Hronov", "type": "location", "pos": [18, 24]}, {"name": "Bohemia", "type": "location", "pos": [27, 34]}, {"name": "Austria-Hungary", "type": "country", "pos": [37, 52]}, {"name": "Czechoslovakia", "type": "country", "pos": [61, 75]}, {"name": "Czech Republic", "type": "country", "pos": [86, 100]}]}, {"sentence": "Some specialized software can narrate RSS .", "entities": [{"name": "RSS", "type": "product", "pos": [38, 41]}]}, {"sentence": "Aspects of ontology editors include : visual navigation possibilities within the knowledge model , inference engine s and extraction ; support for modules ; the import and export of foreign knowledge representation languages for ontology matching ; and the support of meta-ontologies such as OWL-S , Dublin Core , etc .", "entities": [{"name": "visual navigation", "type": "task", "pos": [38, 55]}, {"name": "knowledge model", "type": "task", "pos": [81, 96]}, {"name": "inference engine", "type": "task", "pos": [99, 115]}, {"name": "extraction", "type": "task", "pos": [122, 132]}, {"name": "support for modules", "type": "task", "pos": [135, 154]}, {"name": "knowledge representation", "type": "task", "pos": [190, 214]}, {"name": "ontology matching", "type": "task", "pos": [229, 246]}, {"name": "meta-ontologies", "type": "task", "pos": [268, 283]}, {"name": "OWL-S", "type": "product", "pos": [292, 297]}, {"name": "Dublin Core", "type": "product", "pos": [300, 311]}]}, {"sentence": "The FBI has also instituted its Next Generation Identification program to include face recognition , as well as more traditional biometrics like fingerprints and iris scans , which can pull from both criminal and civil databases .", "entities": [{"name": "FBI", "type": "organization", "pos": [4, 7]}, {"name": "Next Generation Identification program", "type": "else", "pos": [32, 70]}, {"name": "face recognition", "type": "task", "pos": [82, 98]}, {"name": "biometrics", "type": "field", "pos": [129, 139]}, {"name": "fingerprints", "type": "else", "pos": [145, 157]}, {"name": "iris scans", "type": "else", "pos": [162, 172]}]}, {"sentence": "For the 2016 season , Samantha Ponder was added as host , replacing Molly McGrath .", "entities": [{"name": "Samantha Ponder", "type": "person", "pos": [22, 37]}, {"name": "Molly McGrath", "type": "person", "pos": [68, 81]}]}, {"sentence": "It is an adversarial search algorithm used commonly for machine playing of two-player games ( Tic-tac-toe , Chess , Go , etc .", "entities": [{"name": "adversarial search algorithm", "type": "algorithm", "pos": [9, 37]}, {"name": "Tic-tac-toe", "type": "else", "pos": [94, 105]}, {"name": "Chess", "type": "else", "pos": [108, 113]}, {"name": "Go", "type": "else", "pos": [116, 118]}]}, {"sentence": "It involves the fields of computer vision or machine vision , and medical imaging , and makes heavy use of pattern recognition , digital geometry , and signal processing .", "entities": [{"name": "computer vision", "type": "field", "pos": [26, 41]}, {"name": "machine vision", "type": "field", "pos": [45, 59]}, {"name": "medical imaging", "type": "field", "pos": [66, 81]}, {"name": "pattern recognition", "type": "field", "pos": [107, 126]}, {"name": "digital geometry", "type": "field", "pos": [129, 145]}, {"name": "signal processing", "type": "field", "pos": [152, 169]}]}, {"sentence": "In facial recognition system , for instance , a picture of a person 's face would be the input , and the output label would be that person 's name .", "entities": [{"name": "facial recognition system", "type": "product", "pos": [3, 28]}]}, {"sentence": "Apple Inc introduced Face ID on the flagship iPhone X as a biometric authentication successor to the Touch ID , a fingerprint based system .", "entities": [{"name": "Apple Inc", "type": "organization", "pos": [0, 9]}, {"name": "Face ID", "type": "product", "pos": [21, 28]}, {"name": "iPhone X", "type": "product", "pos": [45, 53]}, {"name": "Touch ID", "type": "product", "pos": [101, 109]}]}, {"sentence": "Or combine the F-measure with the R-square evaluated for the raw model output and the target ; or the cost / gain matrix with the correlation coefficient , and so on .", "entities": [{"name": "F-measure", "type": "metrics", "pos": [15, 24]}, {"name": "R-square", "type": "metrics", "pos": [34, 42]}, {"name": "cost / gain matrix", "type": "metrics", "pos": [102, 120]}, {"name": "correlation coefficient", "type": "metrics", "pos": [130, 153]}]}, {"sentence": "The Spanish edition of Campus Party has been held at the Colegio Miguel Hernández , Ceulaj , and the Municipal Sport Arena of Benalmádena in Málaga , Spain ; and at both the Valencia County Fair and the City of Arts and Sciences in Valencia over the past 15 years .", "entities": [{"name": "Spanish edition of Campus Party", "type": "conference", "pos": [4, 35]}, {"name": "Colegio Miguel Hernández", "type": "location", "pos": [57, 81]}, {"name": "Ceulaj", "type": "location", "pos": [84, 90]}, {"name": "Municipal Sport Arena of Benalmádena", "type": "location", "pos": [101, 137]}, {"name": "Málaga", "type": "location", "pos": [141, 147]}, {"name": "Spain", "type": "country", "pos": [150, 155]}, {"name": "Valencia County Fair", "type": "location", "pos": [174, 194]}, {"name": "City of Arts and Sciences", "type": "location", "pos": [203, 228]}, {"name": "Valencia", "type": "location", "pos": [232, 240]}]}, {"sentence": "gnuplot can be used from various programming languages to graph data , including Perl ( via PDL and CPAN packages ) , Python ( via ) .", "entities": [{"name": "gnuplot", "type": "product", "pos": [0, 7]}, {"name": "Perl", "type": "program language", "pos": [81, 85]}, {"name": "PDL", "type": "product", "pos": [92, 95]}, {"name": "CPAN", "type": "product", "pos": [100, 104]}, {"name": "Python", "type": "program language", "pos": [118, 124]}]}, {"sentence": "The field of spoken dialog systems is quite large and includes research ( featured at scientific conferences such as SIGdial and Interspeech ) and a large industrial sector ( with its own meetings such as SpeechTek and AVIOS ) .", "entities": [{"name": "spoken dialog systems", "type": "product", "pos": [13, 34]}, {"name": "SIGdial", "type": "conference", "pos": [117, 124]}, {"name": "Interspeech", "type": "conference", "pos": [129, 140]}, {"name": "SpeechTek", "type": "conference", "pos": [205, 214]}, {"name": "AVIOS", "type": "conference", "pos": [219, 224]}]}, {"sentence": "Challenges in natural language processing frequently involve speech recognition , natural language understanding , and natural language generation .", "entities": [{"name": "natural language processing", "type": "field", "pos": [14, 41]}, {"name": "speech recognition", "type": "task", "pos": [61, 79]}, {"name": "natural language understanding", "type": "task", "pos": [82, 112]}, {"name": "natural language generation", "type": "task", "pos": [119, 146]}]}, {"sentence": "These systems , such as Siri of the iOS operating system , operate on a similar pattern-recognizing technique as that of text-based systems , but with the former , the user input is conducted through speech recognition .", "entities": [{"name": "Siri", "type": "product", "pos": [24, 28]}, {"name": "iOS operating system", "type": "product", "pos": [36, 56]}, {"name": "speech recognition", "type": "task", "pos": [200, 218]}]}, {"sentence": "More exotic fitness functions that explore model granularity include the area under the ROC curve and rank measure .", "entities": [{"name": "exotic fitness functions", "type": "algorithm", "pos": [5, 29]}, {"name": "ROC curve", "type": "metrics", "pos": [88, 97]}]}, {"sentence": "The term Semantic Web was coined by Tim Berners-Lee , the inventor of the World Wide Web and director of the World Wide Web Consortium ( W3C ) , which oversees the development of proposed Semantic Web standards .", "entities": [{"name": "Semantic Web", "type": "product", "pos": [9, 21]}, {"name": "Tim Berners-Lee", "type": "researcher", "pos": [36, 51]}, {"name": "World Wide Web", "type": "product", "pos": [74, 88]}, {"name": "World Wide Web Consortium", "type": "organization", "pos": [109, 134]}, {"name": "W3C", "type": "organization", "pos": [137, 140]}, {"name": "Semantic Web standards", "type": "product", "pos": [188, 210]}]}, {"sentence": "Machine translation , sometimes referred to by the abbreviation MT ( not to be confused with computer-aided translation , machine-aided human translation ( MAHT ) or interactive translation ) , is a sub-field of computational linguistics that investigates the use of software to translate text or speech from one language to another .", "entities": [{"name": "Machine translation", "type": "task", "pos": [0, 19]}, {"name": "MT", "type": "task", "pos": [64, 66]}, {"name": "computer-aided translation", "type": "product", "pos": [93, 119]}, {"name": "machine-aided human translation", "type": "product", "pos": [122, 153]}, {"name": "MAHT", "type": "product", "pos": [156, 160]}, {"name": "interactive translation", "type": "product", "pos": [166, 189]}, {"name": "computational linguistics", "type": "field", "pos": [212, 237]}]}, {"sentence": "Early interlingual MT systems were also built at Stanford in the 1970s by Roger Schank and Yorick Wilks ; the former became the basis of a commercial system for the transfer of funds , and the latter 's code is preserved at The Computer Museum at Boston as the first interlingual machine translation system .", "entities": [{"name": "interlingual MT systems", "type": "product", "pos": [6, 29]}, {"name": "Stanford", "type": "university", "pos": [49, 57]}, {"name": "Roger Schank", "type": "researcher", "pos": [74, 86]}, {"name": "Yorick Wilks", "type": "researcher", "pos": [91, 103]}, {"name": "The Computer Museum", "type": "location", "pos": [224, 243]}, {"name": "Boston", "type": "location", "pos": [247, 253]}, {"name": "interlingual machine translation system", "type": "product", "pos": [267, 306]}]}, {"sentence": "Sycara served as the program chair of the Second International Semantic Web Conference ( ISWC 2003 ) ; general chair , of the Second International Conference on Autonomous Agents ( Agents 98 ) ; chair of the Steering Committee of the Agents Conference ( 1999-2001 ) ; scholarship chair of AAAI ( 1993-1999 ) ;", "entities": [{"name": "Sycara", "type": "researcher", "pos": [0, 6]}, {"name": "Second International Semantic Web Conference", "type": "conference", "pos": [42, 86]}, {"name": "ISWC 2003", "type": "conference", "pos": [89, 98]}, {"name": "Second International Conference on Autonomous Agents", "type": "conference", "pos": [126, 178]}, {"name": "Agents 98", "type": "conference", "pos": [181, 190]}, {"name": "Steering Committee of the Agents Conference", "type": "organization", "pos": [208, 251]}, {"name": "AAAI", "type": "conference", "pos": [289, 293]}]}, {"sentence": "In 2016 , she was selected as the ACL ( Association for Computational Linguistics ) Lifetime Achievement Award winner .", "entities": [{"name": "ACL", "type": "conference", "pos": [34, 37]}, {"name": "Association for Computational Linguistics", "type": "conference", "pos": [40, 81]}, {"name": "Lifetime Achievement Award", "type": "else", "pos": [84, 110]}]}, {"sentence": "Sepp Hochreiter , Y. Bengio , P. Frasconi , and Jürgen Schmidhuber .", "entities": [{"name": "Sepp Hochreiter", "type": "researcher", "pos": [0, 15]}, {"name": "Y. Bengio", "type": "researcher", "pos": [18, 27]}, {"name": "P. Frasconi", "type": "researcher", "pos": [30, 41]}, {"name": "Jürgen Schmidhuber", "type": "researcher", "pos": [48, 66]}]}, {"sentence": "For example , A.L.I.C.E. uses a markup language called AIML , which is specific to its function as a dialogue system , and has since been adopted by various other developers of , so-called , Alicebot s .", "entities": [{"name": "A.L.I.C.E.", "type": "product", "pos": [14, 24]}, {"name": "markup language", "type": "else", "pos": [32, 47]}, {"name": "AIML", "type": "program language", "pos": [55, 59]}, {"name": "dialogue system", "type": "product", "pos": [101, 116]}, {"name": "Alicebot", "type": "product", "pos": [191, 199]}]}, {"sentence": "In 2000 , she was elected as a Fellow of the Association for the Advancement of Artificial Intelligence .", "entities": [{"name": "Association for the Advancement of Artificial Intelligence", "type": "conference", "pos": [45, 103]}]}, {"sentence": "Learning classifier systems ( LCS ) are a family of rule-based machine learning algorithms that combine a discovery component , typically a genetic algorithm , with a learning component , performing either supervised learning , reinforcement learning , or unsupervised learning .", "entities": [{"name": "Learning classifier systems", "type": "else", "pos": [0, 27]}, {"name": "LCS", "type": "else", "pos": [30, 33]}, {"name": "rule-based machine learning algorithms", "type": "else", "pos": [52, 90]}, {"name": "genetic algorithm", "type": "algorithm", "pos": [140, 157]}, {"name": "supervised learning", "type": "field", "pos": [206, 225]}, {"name": "reinforcement learning", "type": "field", "pos": [228, 250]}, {"name": "unsupervised learning", "type": "field", "pos": [256, 277]}]}, {"sentence": "The unknown parameters in each vector βsubk / sub are typically jointly estimated by maximum a posteriori ( MAP ) estimation , which is an extension of maximum likelihood using regularization of the weights to prevent pathological solutions ( usually a squared regularizing function , which is equivalent to placing a zero-mean Gaussian prior distribution on the weights , but other distributions are also possible ) .", "entities": [{"name": "maximum a posteriori", "type": "algorithm", "pos": [85, 105]}, {"name": "MAP", "type": "algorithm", "pos": [108, 111]}, {"name": "maximum likelihood", "type": "algorithm", "pos": [152, 170]}, {"name": "regularization", "type": "else", "pos": [177, 191]}, {"name": "squared regularizing function", "type": "algorithm", "pos": [253, 282]}, {"name": "zero-mean Gaussian prior distribution", "type": "else", "pos": [318, 355]}]}, {"sentence": "The hierarchical structure of words has been explicitly mapped in George Miller ' s Wordnet .", "entities": [{"name": "George Miller", "type": "researcher", "pos": [66, 79]}, {"name": "Wordnet", "type": "product", "pos": [84, 91]}]}, {"sentence": "An illustration of their capabilities is given by the ImageNet Large Scale Visual Recognition Challenge ; this is a benchmark in object classification and detection , with millions of images and hundreds of object classes .", "entities": [{"name": "ImageNet Large Scale Visual Recognition Challenge", "type": "conference", "pos": [54, 103]}, {"name": "object classification and detection", "type": "task", "pos": [129, 164]}]}, {"sentence": "In science fiction , female-appearing robots are often produced for use as domestic servants and sexual slaves , as seen in the film Westworld , Paul J. McAuley ' s novel Fairyland ( 1995 ) , and Lester del Rey ' s short story Helen O 'Loy ( 1938 ) , and sometimes as warriors , killers , or laborers .", "entities": [{"name": "science fiction", "type": "else", "pos": [3, 18]}, {"name": "Westworld", "type": "else", "pos": [133, 142]}, {"name": "Paul J. McAuley", "type": "person", "pos": [145, 160]}, {"name": "Fairyland", "type": "else", "pos": [171, 180]}, {"name": "Lester del Rey", "type": "person", "pos": [196, 210]}, {"name": "Helen O 'Loy", "type": "else", "pos": [227, 239]}]}, {"sentence": "question answering , speech recognition , and machine translation .", "entities": [{"name": "question answering", "type": "task", "pos": [0, 18]}, {"name": "speech recognition", "type": "task", "pos": [21, 39]}, {"name": "machine translation", "type": "task", "pos": [46, 65]}]}, {"sentence": "In his seminal paper , Harry Blum of the Air Force Cambridge Research Laboratories at Hanscom Air Force Base , in Bedford , Massachusetts , defined a medial axis for computing a skeleton of a shape , using an intuitive model of fire propagation on a grass field , where the field has the form of the given shape .", "entities": [{"name": "Harry Blum", "type": "researcher", "pos": [23, 33]}, {"name": "Air Force Cambridge Research Laboratories", "type": "organization", "pos": [41, 82]}, {"name": "Hanscom Air Force Base", "type": "location", "pos": [86, 108]}, {"name": "Bedford", "type": "location", "pos": [114, 121]}, {"name": "Massachusetts", "type": "location", "pos": [124, 137]}]}, {"sentence": "However , in contrast to boosting algorithms that analytically minimize a convex loss function ( e.g. AdaBoost and LogitBoost ) , BrownBoost solves a system of two equations and two unknowns using standard numerical methods .", "entities": [{"name": "AdaBoost", "type": "algorithm", "pos": [102, 110]}, {"name": "LogitBoost", "type": "algorithm", "pos": [115, 125]}, {"name": "BrownBoost", "type": "algorithm", "pos": [130, 140]}]}, {"sentence": "Getoor has multiple best paper awards , an NSF Career Award , and is an Association for the Advancement of Artificial Intelligence ( AAAI ) Fellow .", "entities": [{"name": "Getoor", "type": "researcher", "pos": [0, 6]}, {"name": "NSF Career Award", "type": "else", "pos": [43, 59]}, {"name": "Association for the Advancement of Artificial Intelligence", "type": "conference", "pos": [72, 130]}, {"name": "AAAI", "type": "conference", "pos": [133, 137]}]}, {"sentence": "ACM Fellow ( 2015 ) br Association for Computational Linguistics Fellow ( 2011 ) br AAAI Fellow ( 1994 ) br International Speech Communication Association Fellow ( 2011 ) br Honorary Doctorate ( Hedersdoktor ) KTH Royal Institute of Technology ( 2007 ) br Columbia Engineering School Alumni Association Distinguished Faculty Teaching award ( 2009 ) br IEEE James L. Flanagan Speech and Audio Processing Award ( 2011 ) br ISCA Medal for Scientific Achievement ( 2011 )", "entities": [{"name": "ACM Fellow", "type": "else", "pos": [0, 10]}, {"name": "Association for Computational Linguistics Fellow", "type": "else", "pos": [23, 71]}, {"name": "AAAI Fellow", "type": "else", "pos": [84, 95]}, {"name": "International Speech Communication Association Fellow", "type": "else", "pos": [108, 161]}, {"name": "Honorary Doctorate", "type": "else", "pos": [174, 192]}, {"name": "KTH Royal Institute of Technology", "type": "university", "pos": [210, 243]}, {"name": "Columbia Engineering School Alumni Association Distinguished Faculty Teaching award", "type": "else", "pos": [256, 339]}, {"name": "IEEE James L. Flanagan Speech and Audio Processing Award", "type": "else", "pos": [352, 408]}, {"name": "ISCA Medal for Scientific Achievement", "type": "else", "pos": [421, 458]}]}, {"sentence": "A frustrating outcome of the same study by Stanford ( and other attempts to improve named recognition translation ) is that many times , a decrease in the Bilingual evaluation understudy scores for translation will result from the inclusion of methods for named entity translation .", "entities": [{"name": "Stanford", "type": "university", "pos": [43, 51]}, {"name": "named recognition translation", "type": "task", "pos": [84, 113]}, {"name": "Bilingual evaluation understudy", "type": "metrics", "pos": [155, 186]}, {"name": "named entity translation", "type": "task", "pos": [256, 280]}]}, {"sentence": "Medtronic is using the collected PM data and is working with researchers at Johns Hopkins Hospital and Washington University School of Medicine in order to help answer specific questions about heart disease , such as whether weak hearts cause arrhythmias or vice versa .", "entities": [{"name": "Medtronic", "type": "organization", "pos": [0, 9]}, {"name": "Johns Hopkins Hospital", "type": "organization", "pos": [76, 98]}, {"name": "Washington University School of Medicine", "type": "university", "pos": [103, 143]}]}, {"sentence": "Following that was Paramount 's first feature , Sangaree with Fernando Lamas and Arlene Dahl .", "entities": [{"name": "Paramount", "type": "organization", "pos": [19, 28]}, {"name": "Sangaree", "type": "else", "pos": [48, 56]}, {"name": "Fernando Lamas", "type": "person", "pos": [62, 76]}, {"name": "Arlene Dahl", "type": "person", "pos": [81, 92]}]}, {"sentence": "KRL is a knowledge representation language , developed by Daniel G. Bobrow and Terry Winograd while at Xerox PARC and Stanford University , respectively .", "entities": [{"name": "KRL", "type": "program language", "pos": [0, 3]}, {"name": "Daniel G. Bobrow", "type": "researcher", "pos": [58, 74]}, {"name": "Terry Winograd", "type": "researcher", "pos": [79, 93]}, {"name": "Xerox PARC", "type": "organization", "pos": [103, 113]}, {"name": "Stanford University", "type": "university", "pos": [118, 137]}]}, {"sentence": "At the IEEE Conference on Computer Vision and Pattern Recognition in 2006 , Qiang Zhu , Shai Avidan , Mei-Chen Yeh , and Kwang-Ting Cheng presented an algorithm to significantly speed up human detection using HOG descriptor methods .", "entities": [{"name": "IEEE Conference on Computer Vision and Pattern Recognition", "type": "conference", "pos": [7, 65]}, {"name": "Qiang Zhu", "type": "researcher", "pos": [76, 85]}, {"name": "Shai Avidan", "type": "researcher", "pos": [88, 99]}, {"name": "Mei-Chen Yeh", "type": "researcher", "pos": [102, 114]}, {"name": "Kwang-Ting Cheng", "type": "researcher", "pos": [121, 137]}, {"name": "human detection", "type": "task", "pos": [187, 202]}, {"name": "HOG descriptor methods", "type": "algorithm", "pos": [209, 231]}]}, {"sentence": "Hayes is a charter Fellow of AAAI and of the Cognitive Science Society", "entities": [{"name": "Hayes", "type": "researcher", "pos": [0, 5]}, {"name": "AAAI", "type": "conference", "pos": [29, 33]}]}, {"sentence": "Time series are used in statistics , signal processing , pattern recognition , econometrics , mathematical finance , weather forecasting , earthquake prediction , electroencephalography , control engineering , astronomy , communications engineering , and largely in any domain of applied science and engineering which involves temporal measurements .", "entities": [{"name": "Time series", "type": "else", "pos": [0, 11]}, {"name": "statistics", "type": "field", "pos": [24, 34]}, {"name": "signal processing", "type": "field", "pos": [37, 54]}, {"name": "pattern recognition", "type": "field", "pos": [57, 76]}, {"name": "econometrics", "type": "field", "pos": [79, 91]}, {"name": "mathematical finance", "type": "field", "pos": [94, 114]}, {"name": "weather forecasting", "type": "field", "pos": [117, 136]}, {"name": "earthquake prediction", "type": "field", "pos": [139, 160]}, {"name": "electroencephalography", "type": "field", "pos": [163, 185]}, {"name": "control engineering", "type": "field", "pos": [188, 207]}, {"name": "astronomy", "type": "field", "pos": [210, 219]}, {"name": "communications engineering", "type": "field", "pos": [222, 248]}, {"name": "applied science", "type": "field", "pos": [280, 295]}]}, {"sentence": "In principle , exact recovery can be solved in its feasible range using maximum likelihood , but this amounts to solving a constrained or regularized cut problem such as minimum bisection that is typically NP-complete .", "entities": [{"name": "maximum likelihood", "type": "metrics", "pos": [72, 90]}, {"name": "NP-complete", "type": "else", "pos": [206, 217]}]}, {"sentence": "in their work for pedestrian detection , that was first described at the BMVC in 2009 .", "entities": [{"name": "pedestrian detection", "type": "task", "pos": [18, 38]}, {"name": "BMVC", "type": "conference", "pos": [73, 77]}]}, {"sentence": "In 2007 , at the International Conference on Computer Vision , Terzopoulos was awarded the inaugural IEEE PAMI Computer Vision Distinguished Researcher Award for pioneering and sustained research on deformable models and their applications .", "entities": [{"name": "International Conference on Computer Vision", "type": "conference", "pos": [17, 60]}, {"name": "Terzopoulos", "type": "researcher", "pos": [63, 74]}, {"name": "inaugural IEEE PAMI Computer Vision Distinguished Researcher Award", "type": "else", "pos": [91, 157]}]}, {"sentence": "Cluster analysis or cluster analysis involves assigning data points to clusters such that items in the same cluster are as similar as possible , while items belonging to different clusters are as dissimilar as possible .", "entities": [{"name": "Cluster analysis", "type": "task", "pos": [0, 16]}, {"name": "cluster analysis", "type": "task", "pos": [20, 36]}]}, {"sentence": "( 2005 ) we can differ three different perspectives of text mining , namely text mining as information extraction , text mining as text data mining , and text mining as Data mining ( Knowledge Discovery in Databases ) process.Hotho , A. , Nürnberger , A. and Paaß , G. ( 2005 ) .", "entities": [{"name": "text mining", "type": "field", "pos": [55, 66]}, {"name": "text mining", "type": "field", "pos": [76, 87]}, {"name": "information extraction", "type": "task", "pos": [91, 113]}, {"name": "text mining", "type": "field", "pos": [116, 127]}, {"name": "data mining", "type": "field", "pos": [136, 147]}, {"name": "text mining", "type": "field", "pos": [154, 165]}, {"name": "Data mining", "type": "field", "pos": [169, 180]}, {"name": "Knowledge Discovery", "type": "task", "pos": [183, 202]}, {"name": "Databases", "type": "else", "pos": [206, 215]}]}, {"sentence": "The Rancho Arm was developed as a robotic arm to help handicapped patients at the Rancho Los Amigos National Rehabilitation Center in Downey , California ; this computer-controlled arm was bought by Stanford University in 1963 .", "entities": [{"name": "Rancho Arm", "type": "product", "pos": [4, 14]}, {"name": "Rancho Los Amigos National Rehabilitation Center", "type": "location", "pos": [82, 130]}, {"name": "Downey", "type": "location", "pos": [134, 140]}, {"name": "California", "type": "location", "pos": [143, 153]}, {"name": "Stanford University", "type": "university", "pos": [199, 218]}]}, {"sentence": "At UCSD , Norman was a founder of the Institute for Cognitive Science and one of the organizers of the Cognitive Science Society ( along with Roger Schank , Allan M. Collins , and others ) , which held its first meeting at the UCSD campus in 1979 .", "entities": [{"name": "UCSD", "type": "university", "pos": [3, 7]}, {"name": "Norman", "type": "researcher", "pos": [10, 16]}, {"name": "the Institute for Cognitive Science", "type": "organization", "pos": [34, 69]}, {"name": "Cognitive Science Society", "type": "organization", "pos": [103, 128]}, {"name": "Roger Schank", "type": "researcher", "pos": [142, 154]}, {"name": "Allan M. Collins", "type": "researcher", "pos": [157, 173]}, {"name": "UCSD", "type": "university", "pos": [227, 231]}]}, {"sentence": "The most commonly used robot configurations are articulated robots , SCARA robots , delta robots and cartesian coordinate robots , ( gantry robots or x-y-z robots ) .", "entities": [{"name": "articulated robots", "type": "product", "pos": [48, 66]}, {"name": "SCARA robots", "type": "product", "pos": [69, 81]}, {"name": "delta robots", "type": "product", "pos": [84, 96]}, {"name": "cartesian coordinate robots", "type": "product", "pos": [101, 128]}, {"name": "gantry robots", "type": "product", "pos": [133, 146]}, {"name": "x-y-z robots", "type": "product", "pos": [150, 162]}]}, {"sentence": "Alternatively , it can be used directly with the Perl Module TM ( which also supports LTM ) .", "entities": [{"name": "Perl", "type": "program language", "pos": [49, 53]}, {"name": "Module TM", "type": "else", "pos": [54, 63]}, {"name": "LTM", "type": "else", "pos": [86, 89]}]}, {"sentence": "This was won by an United States team from Newton Labs , and the competition was shown on CNN .", "entities": [{"name": "United States", "type": "country", "pos": [19, 32]}, {"name": "Newton Labs", "type": "organization", "pos": [43, 54]}, {"name": "CNN", "type": "organization", "pos": [90, 93]}]}, {"sentence": "The Butler 's in Love , a short film directed by David Arquette and starring Elizabeth Berkley and Thomas Jane was released on June 23 , 2008 .", "entities": [{"name": "The Butler 's in Love", "type": "else", "pos": [0, 21]}, {"name": "David Arquette", "type": "person", "pos": [49, 63]}, {"name": "Elizabeth Berkley", "type": "person", "pos": [77, 94]}, {"name": "Thomas Jane", "type": "person", "pos": [99, 110]}]}, {"sentence": "For instance , WordNet is a resource including a taxonomy , whose elements are meanings of English words .", "entities": [{"name": "WordNet", "type": "product", "pos": [15, 22]}, {"name": "taxonomy", "type": "field", "pos": [49, 57]}, {"name": "English", "type": "else", "pos": [91, 98]}]}, {"sentence": "Existing humanoid robot systems such as ASIMO and QRIO use many motors to achieve locomotion .", "entities": [{"name": "humanoid robot systems", "type": "product", "pos": [9, 31]}, {"name": "ASIMO", "type": "product", "pos": [40, 45]}, {"name": "QRIO", "type": "product", "pos": [50, 54]}, {"name": "locomotion", "type": "else", "pos": [82, 92]}]}, {"sentence": "LEPOR is designed with the factors of enhanced length penalty , precision , n-gram word order penalty , and recall .", "entities": [{"name": "LEPOR", "type": "metrics", "pos": [0, 5]}, {"name": "precision", "type": "metrics", "pos": [64, 73]}, {"name": "n-gram word order penalty", "type": "else", "pos": [76, 101]}, {"name": "recall", "type": "metrics", "pos": [108, 114]}]}, {"sentence": "It is based on the Bilingual evaluation understudy metric , but with some alterations .", "entities": [{"name": "Bilingual evaluation understudy metric", "type": "metrics", "pos": [19, 57]}]}, {"sentence": "This is an example implementation in MATLAB / Octave :", "entities": [{"name": "MATLAB", "type": "product", "pos": [37, 43]}, {"name": "Octave", "type": "product", "pos": [46, 52]}]}, {"sentence": "It is designed to be used through a number of computer languages , include Python , Ruby , and Scheme .", "entities": [{"name": "Python", "type": "program language", "pos": [75, 81]}, {"name": "Ruby", "type": "program language", "pos": [84, 88]}, {"name": "Scheme", "type": "program language", "pos": [95, 101]}]}, {"sentence": "Hayes has served as secretary of AISB , chairman and trustee of IJCAI , associate editor of Artificial Intelligence , a governor of the Cognitive Science Society and president of American Association for Artificial Intelligence .", "entities": [{"name": "Hayes", "type": "researcher", "pos": [0, 5]}, {"name": "AISB", "type": "organization", "pos": [33, 37]}, {"name": "IJCAI", "type": "conference", "pos": [64, 69]}, {"name": "Artificial Intelligence", "type": "field", "pos": [92, 115]}, {"name": "Cognitive Science Society", "type": "organization", "pos": [136, 161]}, {"name": "American Association for Artificial Intelligence", "type": "organization", "pos": [179, 227]}]}, {"sentence": "Two of them , Now is the Time ( to Put On Your Glasses ) and Around is Around , were directed by Norman McLaren in 1951 for the National Film Board of Canada .", "entities": [{"name": "Now is the Time ( to Put On Your Glasses )", "type": "else", "pos": [14, 56]}, {"name": "Around is Around", "type": "else", "pos": [61, 77]}, {"name": "Norman McLaren", "type": "person", "pos": [97, 111]}, {"name": "National Film Board of Canada", "type": "organization", "pos": [128, 157]}]}, {"sentence": "A recommender system aims to predict the preference for an item of a target user .", "entities": [{"name": "recommender system", "type": "product", "pos": [2, 20]}]}, {"sentence": "Convolution has applications that include probability , statistics , computer vision , natural language processing , image processing and signal processing , engineering , and differential equations .", "entities": [{"name": "Convolution", "type": "algorithm", "pos": [0, 11]}, {"name": "probability", "type": "field", "pos": [42, 53]}, {"name": "statistics", "type": "field", "pos": [56, 66]}, {"name": "computer vision", "type": "field", "pos": [69, 84]}, {"name": "natural language processing", "type": "field", "pos": [87, 114]}, {"name": "image processing", "type": "field", "pos": [117, 133]}, {"name": "signal processing", "type": "field", "pos": [138, 155]}, {"name": "engineering", "type": "field", "pos": [158, 169]}, {"name": "differential equations", "type": "field", "pos": [176, 198]}]}, {"sentence": "Applications of DSP include audio signal processing , audio compression , digital image processing , video compression , speech processing , speech recognition , digital communication s , digital synthesizer s , radar , sonar , financial signal processing , seismology and biomedicine .", "entities": [{"name": "DSP", "type": "field", "pos": [16, 19]}, {"name": "audio signal processing", "type": "task", "pos": [28, 51]}, {"name": "audio compression", "type": "task", "pos": [54, 71]}, {"name": "digital", "type": "task", "pos": [74, 81]}, {"name": "image", "type": "task", "pos": [82, 87]}, {"name": "processing", "type": "task", "pos": [88, 98]}, {"name": "video compression", "type": "task", "pos": [101, 118]}, {"name": "speech processing", "type": "task", "pos": [121, 138]}, {"name": "speech recognition", "type": "task", "pos": [141, 159]}, {"name": "digital communication", "type": "task", "pos": [162, 183]}, {"name": "digital synthesizer", "type": "task", "pos": [188, 207]}, {"name": "radar", "type": "field", "pos": [212, 217]}, {"name": "sonar", "type": "field", "pos": [220, 225]}, {"name": "financial signal processing", "type": "field", "pos": [228, 255]}, {"name": "seismology", "type": "field", "pos": [258, 268]}, {"name": "biomedicine", "type": "field", "pos": [273, 284]}]}, {"sentence": "( February 20 , 1912 - August 11 , 2011 ) was an American inventor , best known for creating Unimate , the first industrial robot .", "entities": [{"name": "American", "type": "else", "pos": [49, 57]}, {"name": "Unimate", "type": "product", "pos": [93, 100]}]}, {"sentence": "With David E. Rumelhart and Ronald J. Williams , Hinton was co-author of a highly cited paper published in 1986 that popularized the backpropagation algorithm for training multi-layer neural networks , The dramatic image-recognition milestone of the AlexNet designed by his student Alex Krizhevsky { { cite web", "entities": [{"name": "David E. Rumelhart", "type": "researcher", "pos": [5, 23]}, {"name": "Ronald J. Williams", "type": "researcher", "pos": [28, 46]}, {"name": "Hinton", "type": "researcher", "pos": [49, 55]}, {"name": "backpropagation algorithm", "type": "algorithm", "pos": [133, 158]}, {"name": "multi-layer neural networks", "type": "algorithm", "pos": [172, 199]}, {"name": "image-recognition", "type": "task", "pos": [215, 232]}, {"name": "AlexNet", "type": "algorithm", "pos": [250, 257]}, {"name": "Alex Krizhevsky", "type": "researcher", "pos": [282, 297]}]}, {"sentence": "When the value being predicted is continuously distributed , the mean squared error , root mean squared error or median absolute deviation could be used to summarize the errors .", "entities": [{"name": "mean squared error", "type": "metrics", "pos": [65, 83]}, {"name": "root mean squared error", "type": "metrics", "pos": [86, 109]}, {"name": "median absolute deviation", "type": "metrics", "pos": [113, 138]}]}, {"sentence": "Conceptual clustering developed mainly during the 1980s , as a machine learning paradigm for unsupervised learning .", "entities": [{"name": "Conceptual clustering", "type": "algorithm", "pos": [0, 21]}, {"name": "machine learning", "type": "field", "pos": [63, 79]}, {"name": "unsupervised learning", "type": "field", "pos": [93, 114]}]}, {"sentence": "If named entities cannot be recognized by the machine translator , they may be erroneously translated as common nouns , which would most likely not affect the Bilingual evaluation understudy rating of the translation but would change the text 's human readability .", "entities": [{"name": "machine translator", "type": "product", "pos": [46, 64]}, {"name": "Bilingual evaluation understudy", "type": "metrics", "pos": [159, 190]}]}, {"sentence": "Roger Schank , 1969 , A conceptual dependency parser for natural language Proceedings of the 1969 on Computational linguistics , Sång-Säby , Sweden , pages 1-3 This model , partially influenced by the work of Sydney Lamb , was extensively used by Schank 's students at Yale University , such as Robert Wilensky , Wendy Lehnert , and Janet Kolodner .", "entities": [{"name": "Roger Schank", "type": "researcher", "pos": [0, 12]}, {"name": "natural language", "type": "else", "pos": [57, 73]}, {"name": "Proceedings of the 1969 on Computational linguistics", "type": "conference", "pos": [74, 126]}, {"name": "Sång-Säby", "type": "location", "pos": [129, 138]}, {"name": "Sweden", "type": "country", "pos": [141, 147]}, {"name": "Sydney Lamb", "type": "researcher", "pos": [209, 220]}, {"name": "Schank", "type": "researcher", "pos": [247, 253]}, {"name": "Yale University", "type": "university", "pos": [269, 284]}, {"name": "Robert Wilensky", "type": "researcher", "pos": [295, 310]}, {"name": "Wendy Lehnert", "type": "researcher", "pos": [313, 326]}, {"name": "Janet Kolodner", "type": "researcher", "pos": [333, 347]}]}, {"sentence": "Improved maximum likelihood method ( IMLM ) is a combination of two MLM ( maximum likelihood ) estimators .", "entities": [{"name": "maximum likelihood method", "type": "algorithm", "pos": [9, 34]}, {"name": "IMLM", "type": "algorithm", "pos": [37, 41]}, {"name": "MLM", "type": "algorithm", "pos": [68, 71]}, {"name": "maximum likelihood", "type": "metrics", "pos": [74, 92]}]}, {"sentence": "These methods may also analyze a program 's output and its usefulness and therefore may involve the analysis of its confusion matrix ( or table of confusion ) .", "entities": [{"name": "confusion matrix", "type": "metrics", "pos": [116, 132]}, {"name": "table of confusion", "type": "metrics", "pos": [138, 156]}]}, {"sentence": "SURF was first published by Herbert Bay , Tinne Tuytelaars , and Luc Van Gool , and presented at the 2006 European Conference on Computer Vision .", "entities": [{"name": "SURF", "type": "product", "pos": [0, 4]}, {"name": "Herbert Bay", "type": "researcher", "pos": [28, 39]}, {"name": "Tinne Tuytelaars", "type": "researcher", "pos": [42, 58]}, {"name": "Luc Van Gool", "type": "researcher", "pos": [65, 77]}, {"name": "2006 European Conference on Computer Vision", "type": "conference", "pos": [101, 144]}]}, {"sentence": "OCR is a field of research in pattern recognition , artificial intelligence and computer vision .", "entities": [{"name": "OCR", "type": "task", "pos": [0, 3]}, {"name": "pattern recognition", "type": "field", "pos": [30, 49]}, {"name": "artificial intelligence", "type": "field", "pos": [52, 75]}, {"name": "computer vision", "type": "field", "pos": [80, 95]}]}, {"sentence": "Continuing the example using the maximum likelihood estimator , the probability density function ( pdf ) of the noise for one sample mathw n / math is", "entities": [{"name": "maximum likelihood estimator", "type": "metrics", "pos": [33, 61]}, {"name": "probability density function", "type": "algorithm", "pos": [68, 96]}, {"name": "pdf", "type": "algorithm", "pos": [99, 102]}]}, {"sentence": "Sub-domains of computer vision include scene reconstruction , event detection , video tracking , object recognition , 3D pose estimation , learning , indexing , motion estimation , visual servoing , 3D scene modeling , and image restoration .", "entities": [{"name": "computer vision", "type": "field", "pos": [15, 30]}, {"name": "scene reconstruction", "type": "task", "pos": [39, 59]}, {"name": "event detection", "type": "task", "pos": [62, 77]}, {"name": "video tracking", "type": "task", "pos": [80, 94]}, {"name": "object recognition", "type": "task", "pos": [97, 115]}, {"name": "3D pose estimation", "type": "task", "pos": [118, 136]}, {"name": "learning", "type": "task", "pos": [139, 147]}, {"name": "indexing", "type": "task", "pos": [150, 158]}, {"name": "motion estimation", "type": "task", "pos": [161, 178]}, {"name": "visual servoing", "type": "task", "pos": [181, 196]}, {"name": "3D scene modeling", "type": "task", "pos": [199, 216]}, {"name": "image restoration", "type": "task", "pos": [223, 240]}]}, {"sentence": "In 2013 , at the International Conference on Computer Vision , Terzopoulos was awarded a Helmholtz Prize for his 1987 ICCV paper with Kass and Witkin on active contour model s .", "entities": [{"name": "International Conference on Computer Vision", "type": "conference", "pos": [17, 60]}, {"name": "Terzopoulos", "type": "researcher", "pos": [63, 74]}, {"name": "Helmholtz Prize", "type": "else", "pos": [89, 104]}, {"name": "1987 ICCV", "type": "conference", "pos": [113, 122]}, {"name": "Kass", "type": "researcher", "pos": [134, 138]}, {"name": "Witkin", "type": "researcher", "pos": [143, 149]}, {"name": "active contour model", "type": "algorithm", "pos": [153, 173]}]}, {"sentence": "If the regularization function Many algorithms exist for solving such problems ; popular ones for linear classification include Stochastic gradient descent ) gradient descent , L-BFGS , coordinate descent and Newton method s .", "entities": [{"name": "linear classification", "type": "task", "pos": [98, 119]}, {"name": "Stochastic gradient descent", "type": "algorithm", "pos": [128, 155]}, {"name": "gradient descent", "type": "algorithm", "pos": [158, 174]}, {"name": "L-BFGS", "type": "algorithm", "pos": [177, 183]}, {"name": "coordinate descent", "type": "algorithm", "pos": [186, 204]}, {"name": "Newton method", "type": "algorithm", "pos": [209, 222]}]}, {"sentence": "Long short-term memory ( LSTM ) networks were invented by Sepp Hochreiter and Jürgen Schmidhuber in 1997 and set accuracy records in multiple applications domains .", "entities": [{"name": "Long short-term memory", "type": "algorithm", "pos": [0, 22]}, {"name": "LSTM", "type": "algorithm", "pos": [25, 29]}, {"name": "Sepp Hochreiter", "type": "researcher", "pos": [58, 73]}, {"name": "Jürgen Schmidhuber", "type": "researcher", "pos": [78, 96]}]}, {"sentence": "TN was developed at Massachusetts General Hospital and was tested in multiple scenarios including the extraction of smoking status , family history of coronary artery disease , identifying patients with sleep disorders ,", "entities": [{"name": "TN", "type": "product", "pos": [0, 2]}, {"name": "Massachusetts General Hospital", "type": "organization", "pos": [20, 50]}]}, {"sentence": "In 1960 , Devol personally sold the first Unimate robot , which was shipped in 1961 to General Motors .", "entities": [{"name": "Devol", "type": "researcher", "pos": [10, 15]}, {"name": "Unimate", "type": "product", "pos": [42, 49]}, {"name": "General Motors", "type": "organization", "pos": [87, 101]}]}, {"sentence": "The Campus Party Europe was held April 14-18 , 2010 at the Caja Mágica in Madrid , Spain with 800 participants from each of the 27 European Union member states .", "entities": [{"name": "Campus Party Europe", "type": "conference", "pos": [4, 23]}, {"name": "Caja Mágica", "type": "location", "pos": [59, 70]}, {"name": "Madrid", "type": "location", "pos": [74, 80]}, {"name": "Spain", "type": "country", "pos": [83, 88]}, {"name": "European Union", "type": "organization", "pos": [131, 145]}]}, {"sentence": "In July 2016 , a collaboration between DeepMind and Moorfields Eye Hospital was announced to develop AI applications for healthcare .", "entities": [{"name": "DeepMind", "type": "organization", "pos": [39, 47]}, {"name": "Moorfields Eye Hospital", "type": "organization", "pos": [52, 75]}, {"name": "AI applications for healthcare", "type": "else", "pos": [101, 131]}]}, {"sentence": "They ended up awarding eleven PR2s to different institutions , including University of Freiburg , Bosch , Georgia Tech , KU Leuven , MIT , Stanford , Technical University of Munich , UC Berkeley , U Penn , USC , and University of Tokyo .", "entities": [{"name": "PR2s", "type": "else", "pos": [30, 34]}, {"name": "University of Freiburg", "type": "university", "pos": [73, 95]}, {"name": "Bosch", "type": "university", "pos": [98, 103]}, {"name": "Georgia Tech", "type": "university", "pos": [106, 118]}, {"name": "KU Leuven", "type": "university", "pos": [121, 130]}, {"name": "MIT", "type": "university", "pos": [133, 136]}, {"name": "Stanford", "type": "university", "pos": [139, 147]}, {"name": "Technical University of Munich", "type": "university", "pos": [150, 180]}, {"name": "UC Berkeley", "type": "university", "pos": [183, 194]}, {"name": "U Penn", "type": "university", "pos": [197, 203]}, {"name": "USC", "type": "university", "pos": [206, 209]}, {"name": "University of Tokyo", "type": "university", "pos": [216, 235]}]}, {"sentence": "The counts of TP , TN , FP , and FN are usually kept on a table known as the confusion matrix .", "entities": [{"name": "TP", "type": "metrics", "pos": [14, 16]}, {"name": "TN", "type": "metrics", "pos": [19, 21]}, {"name": "FP", "type": "metrics", "pos": [24, 26]}, {"name": "FN", "type": "metrics", "pos": [33, 35]}, {"name": "confusion matrix", "type": "metrics", "pos": [77, 93]}]}, {"sentence": "As feature set , information gain , cross entropy , mutual information , and odds ratio are usually used .", "entities": [{"name": "information gain", "type": "metrics", "pos": [17, 33]}, {"name": "cross entropy", "type": "metrics", "pos": [36, 49]}, {"name": "mutual information", "type": "metrics", "pos": [52, 70]}, {"name": "odds ratio", "type": "metrics", "pos": [77, 87]}]}, {"sentence": "It has been applied successfully to various problems , including robot control , elevator scheduling , telecommunications , , checkers and Go ( AlphaGo ) .", "entities": [{"name": "robot control", "type": "task", "pos": [65, 78]}, {"name": "elevator scheduling", "type": "task", "pos": [81, 100]}, {"name": "telecommunications", "type": "task", "pos": [103, 121]}, {"name": "checkers", "type": "task", "pos": [126, 134]}, {"name": "Go", "type": "task", "pos": [139, 141]}, {"name": "AlphaGo", "type": "product", "pos": [144, 151]}]}, {"sentence": "In 2018 , the inaugural year of mission 8 , the American Venue was held on the campus of the Georgia Institute of Technology in Atlanta , Georgia , and the Asia / Pacific Venue was conducted at Beihang University Gymnasium in Beijing China .", "entities": [{"name": "American", "type": "else", "pos": [48, 56]}, {"name": "Georgia Institute of Technology", "type": "university", "pos": [93, 124]}, {"name": "Atlanta", "type": "location", "pos": [128, 135]}, {"name": "Georgia", "type": "location", "pos": [138, 145]}, {"name": "Asia / Pacific", "type": "location", "pos": [156, 170]}, {"name": "Beihang University Gymnasium", "type": "location", "pos": [194, 222]}, {"name": "Beijing", "type": "location", "pos": [226, 233]}, {"name": "China", "type": "country", "pos": [234, 239]}]}, {"sentence": "Machine learning is strongly related to pattern recognition and originates from artificial intelligence .", "entities": [{"name": "Machine learning", "type": "field", "pos": [0, 16]}, {"name": "pattern recognition", "type": "field", "pos": [40, 59]}, {"name": "artificial intelligence", "type": "field", "pos": [80, 103]}]}, {"sentence": "It comes with 3 Java games that are controlled with the remote control and displayed to its LCD screen .", "entities": [{"name": "Java", "type": "program language", "pos": [16, 20]}, {"name": "LCD screen", "type": "product", "pos": [92, 102]}]}, {"sentence": "A commercially successful but specialized computer vision-based articulated body pose estimation technique is optical motion capture .", "entities": [{"name": "computer vision-based articulated body pose estimation", "type": "task", "pos": [42, 96]}, {"name": "optical motion capture", "type": "algorithm", "pos": [110, 132]}]}, {"sentence": "The SMC is very similar to the more popular Jaccard index .", "entities": [{"name": "SMC", "type": "organization", "pos": [4, 7]}, {"name": "Jaccard index", "type": "metrics", "pos": [44, 57]}]}, {"sentence": "The PUMA ( Programmable Universal Machine for Assembly , or Programmable Universal Manipulation Arm ) is an industrial robot robotic arm developed by Victor Scheinman at pioneering robot company Unimation .", "entities": [{"name": "PUMA", "type": "product", "pos": [4, 8]}, {"name": "Programmable Universal Machine for Assembly", "type": "product", "pos": [11, 54]}, {"name": "Programmable Universal Manipulation Arm", "type": "product", "pos": [60, 99]}, {"name": "Victor Scheinman", "type": "researcher", "pos": [150, 166]}, {"name": "Unimation", "type": "organization", "pos": [195, 204]}]}, {"sentence": "It is written in Python .", "entities": [{"name": "Python", "type": "program language", "pos": [17, 23]}]}, {"sentence": "Bandwidth in hertz is a central concept in many fields , including electronics , information theory , digital communication s , radio communication s , signal processing , and spectroscopy and is one of the determinants of the capacity of a given communication channel .", "entities": [{"name": "Bandwidth", "type": "else", "pos": [0, 9]}, {"name": "hertz", "type": "else", "pos": [13, 18]}, {"name": "electronics", "type": "field", "pos": [67, 78]}, {"name": "information theory", "type": "field", "pos": [81, 99]}, {"name": "digital communication", "type": "field", "pos": [102, 123]}, {"name": "radio communication", "type": "field", "pos": [128, 147]}, {"name": "signal processing", "type": "field", "pos": [152, 169]}, {"name": "spectroscopy", "type": "field", "pos": [176, 188]}]}, {"sentence": "If a convex loss is utilized ( as in AdaBoost , LogitBoost , and all members of the AnyBoost family of algorithms ) then an example with higher margin will receive less ( or equal ) weight than an example with lower margin .", "entities": [{"name": "AdaBoost", "type": "algorithm", "pos": [37, 45]}, {"name": "LogitBoost", "type": "algorithm", "pos": [48, 58]}, {"name": "AnyBoost family of algorithms", "type": "else", "pos": [84, 113]}]}, {"sentence": "Sepp Hochreiter ' s diploma thesis of 1991 Sepp Hochreiter .", "entities": [{"name": "Sepp Hochreiter", "type": "researcher", "pos": [0, 15]}, {"name": "Sepp Hochreiter", "type": "researcher", "pos": [43, 58]}]}, {"sentence": "Typical discriminative models include logistic regression ( LR ) , support vector machine s ( SVM ) , conditional random fields ( CRFs ) ( specified over an undirected graph ) , decision trees , neural networks , and many others .", "entities": [{"name": "logistic regression", "type": "algorithm", "pos": [38, 57]}, {"name": "LR", "type": "algorithm", "pos": [60, 62]}, {"name": "support vector machine", "type": "algorithm", "pos": [67, 89]}, {"name": "SVM", "type": "algorithm", "pos": [94, 97]}, {"name": "conditional random fields", "type": "algorithm", "pos": [102, 127]}, {"name": "CRFs", "type": "algorithm", "pos": [130, 134]}, {"name": "undirected graph", "type": "algorithm", "pos": [157, 173]}, {"name": "decision trees", "type": "algorithm", "pos": [178, 192]}, {"name": "neural networks", "type": "algorithm", "pos": [195, 210]}]}, {"sentence": "Then it is also possible to use these probabilities and evaluate the mean squared error ( or some other similar measure ) between the probabilities and the actual values , then combine this with the confusion matrix to create very efficient fitness functions for logistic regression .", "entities": [{"name": "mean squared error", "type": "metrics", "pos": [69, 87]}, {"name": "confusion matrix", "type": "metrics", "pos": [199, 215]}]}, {"sentence": "VoiceOver was for the first time featured in 2005 in Mac OS X Tiger ( 10.4 ) .", "entities": [{"name": "VoiceOver", "type": "product", "pos": [0, 9]}, {"name": "Mac OS X Tiger", "type": "product", "pos": [53, 67]}]}, {"sentence": "In practice , machine learning algorithms cope with that either by employing a convex approximation to the 0-1 loss function ( like hinge loss for Support vector machine ) , which is easier to optimize , or by imposing assumptions on the distribution mathP ( x , y ) / math ( and thus stop being agnostic learning algorithms to which the above result applies ) .", "entities": [{"name": "convex approximation", "type": "algorithm", "pos": [79, 99]}, {"name": "loss function", "type": "else", "pos": [111, 124]}, {"name": "hinge loss", "type": "metrics", "pos": [132, 142]}, {"name": "Support vector machine", "type": "algorithm", "pos": [147, 169]}, {"name": "agnostic learning algorithms", "type": "else", "pos": [296, 324]}]}, {"sentence": "Westworld ( 1973 ) was the first feature film to use the digital image processing to photography to simulate an android 's point of view .", "entities": [{"name": "Westworld", "type": "else", "pos": [0, 9]}, {"name": "digital image processing", "type": "field", "pos": [57, 81]}, {"name": "android", "type": "product", "pos": [112, 119]}]}, {"sentence": "It is now also commonly used in speech recognition , speech synthesis , diarization , Xavier Anguera et al .", "entities": [{"name": "speech recognition", "type": "task", "pos": [32, 50]}, {"name": "speech synthesis", "type": "task", "pos": [53, 69]}, {"name": "diarization", "type": "task", "pos": [72, 83]}, {"name": "Xavier Anguera", "type": "researcher", "pos": [86, 100]}]}, {"sentence": "Here , math \\ sigma / math is an element-wise activation function such as a sigmoid function or a rectified linear unit .", "entities": [{"name": "element-wise activation function", "type": "algorithm", "pos": [33, 65]}, {"name": "sigmoid function", "type": "algorithm", "pos": [76, 92]}, {"name": "rectified linear unit", "type": "algorithm", "pos": [98, 119]}]}, {"sentence": "Traditional phonetic-based ( i.e. , all Hidden Markov model -based model ) approaches required separate components and training for the pronunciation , acoustic and language model .", "entities": [{"name": "Hidden Markov model", "type": "algorithm", "pos": [40, 59]}, {"name": "pronunciation", "type": "else", "pos": [136, 149]}, {"name": "acoustic", "type": "else", "pos": [152, 160]}, {"name": "language model", "type": "else", "pos": [165, 179]}]}, {"sentence": "The Roberts cross operator is used in image processing and computer vision for edge detection .", "entities": [{"name": "Roberts cross operator", "type": "algorithm", "pos": [4, 26]}, {"name": "image processing", "type": "field", "pos": [38, 54]}, {"name": "computer vision", "type": "field", "pos": [59, 74]}, {"name": "edge detection", "type": "task", "pos": [79, 93]}]}, {"sentence": "The values of sensitivity and specificity are agnostic to the percent of positive cases in the population of interest ( as opposed to , for example , precision ) .", "entities": [{"name": "sensitivity", "type": "metrics", "pos": [14, 25]}, {"name": "specificity", "type": "metrics", "pos": [30, 41]}, {"name": "precision", "type": "metrics", "pos": [150, 159]}]}, {"sentence": "But perceptron models were made very unpopular by the book Perceptrons by Marvin Minsky and Seymour Papert , published in 1969 .", "entities": [{"name": "perceptron models", "type": "algorithm", "pos": [4, 21]}, {"name": "Perceptrons", "type": "else", "pos": [59, 70]}, {"name": "Marvin Minsky", "type": "researcher", "pos": [74, 87]}, {"name": "Seymour Papert", "type": "researcher", "pos": [92, 106]}]}, {"sentence": "The Document Understanding Conferences , conducted annually by NIST , have developed sophisticated evaluation criteria for techniques accepting the multi-document summarization challenge .", "entities": [{"name": "Document Understanding Conferences", "type": "conference", "pos": [4, 38]}, {"name": "NIST", "type": "organization", "pos": [63, 67]}, {"name": "multi-document summarization", "type": "task", "pos": [148, 176]}]}, {"sentence": "A parallel manipulator is designed so that each chain is usually short , simple and can thus be rigid against unwanted movement , compared to a serial manipulator .", "entities": [{"name": "parallel manipulator", "type": "product", "pos": [2, 22]}, {"name": "serial manipulator", "type": "product", "pos": [144, 162]}]}, {"sentence": "The manipulator is what makes the robot move , and the design of these systems can be categorized into several common types , such as SCARA and cartesian coordinate robot , which use different coordinate systems to direct the arms of the machine .", "entities": [{"name": "SCARA", "type": "else", "pos": [134, 139]}, {"name": "cartesian coordinate robot", "type": "else", "pos": [144, 170]}]}, {"sentence": "In the United States he is a Member of the National Academy of Sciences , the American Academy of Arts and Sciences , the Linguistic Society of America , the American Philosophical Association , and the American Association for the Advancement of Science .", "entities": [{"name": "United States", "type": "country", "pos": [7, 20]}, {"name": "National Academy of Sciences", "type": "organization", "pos": [43, 71]}, {"name": "American Academy of Arts and Sciences", "type": "organization", "pos": [78, 115]}, {"name": "Linguistic Society of America", "type": "organization", "pos": [122, 151]}, {"name": "American Philosophical Association", "type": "organization", "pos": [158, 192]}, {"name": "American Association for the Advancement of Science", "type": "organization", "pos": [203, 254]}]}, {"sentence": "They rose to great prominence with the popularity of the support vector machine ( SVM ) in the 1990s , when the SVM was found to be competitive with neural networks on tasks such as handwriting recognition .", "entities": [{"name": "support vector machine", "type": "algorithm", "pos": [57, 79]}, {"name": "SVM", "type": "algorithm", "pos": [82, 85]}, {"name": "SVM", "type": "algorithm", "pos": [112, 115]}, {"name": "neural networks", "type": "algorithm", "pos": [149, 164]}, {"name": "handwriting recognition", "type": "task", "pos": [182, 205]}]}, {"sentence": "An empirical whitening transform is obtained by estimating the covariance ( e.g. by maximum likelihood ) and subsequently constructing a corresponding estimated whitening matrix ( e.g. by Cholesky decomposition ) .", "entities": [{"name": "whitening transform", "type": "else", "pos": [13, 32]}, {"name": "covariance", "type": "else", "pos": [63, 73]}, {"name": "maximum likelihood", "type": "algorithm", "pos": [84, 102]}, {"name": "whitening matrix", "type": "else", "pos": [161, 177]}, {"name": "Cholesky decomposition", "type": "algorithm", "pos": [188, 210]}]}, {"sentence": "IAI is the world 's largest manufacturer of cartesian coordinate robot s and is an established leader in low cost , high performance SCARA robot s .", "entities": [{"name": "IAI", "type": "organization", "pos": [0, 3]}, {"name": "cartesian coordinate robot", "type": "product", "pos": [44, 70]}, {"name": "SCARA robot", "type": "product", "pos": [133, 144]}]}, {"sentence": "Formal concept analysis finds practical application in fields including data mining , text mining , machine learning , knowledge management , semantic web , software development , chemistry and biology .", "entities": [{"name": "data mining", "type": "field", "pos": [72, 83]}, {"name": "text mining", "type": "field", "pos": [86, 97]}, {"name": "machine learning", "type": "field", "pos": [100, 116]}, {"name": "knowledge management", "type": "field", "pos": [119, 139]}, {"name": "semantic web", "type": "field", "pos": [142, 154]}, {"name": "software development", "type": "field", "pos": [157, 177]}, {"name": "chemistry", "type": "field", "pos": [180, 189]}, {"name": "biology", "type": "field", "pos": [194, 201]}]}, {"sentence": "In computer science , computational learning theory ( or just learning theory ) is a subfield of artificial intelligence devoted to studying the design and analysis of machine learning algorithms .", "entities": [{"name": "computer science", "type": "field", "pos": [3, 19]}, {"name": "computational learning theory", "type": "field", "pos": [22, 51]}, {"name": "learning theory", "type": "field", "pos": [62, 77]}, {"name": "artificial intelligence", "type": "field", "pos": [97, 120]}, {"name": "machine learning", "type": "field", "pos": [168, 184]}]}, {"sentence": "Collaborative filtering ( CF ) is a technique used by recommender system s .", "entities": [{"name": "Collaborative filtering", "type": "algorithm", "pos": [0, 23]}, {"name": "CF", "type": "algorithm", "pos": [26, 28]}, {"name": "recommender system", "type": "product", "pos": [54, 72]}]}, {"sentence": "The FALSE positive rate is the proportion of all negatives that still yield positive test outcomes , i.e. , the conditional probability of a positive test result given an event that was not present .", "entities": [{"name": "FALSE positive rate", "type": "metrics", "pos": [4, 23]}]}, {"sentence": "In VLDB ' 8 : Proceedings of the 34th International Conference on Very Large Data Bases , pages 422--433. showed that the given values for mathC / math and mathK / math generally imply relatively low accuracy of iteratively computed SimRank scores .", "entities": [{"name": "VLDB", "type": "conference", "pos": [3, 7]}, {"name": "Proceedings of the 34th International Conference on Very Large Data Bases", "type": "conference", "pos": [14, 87]}, {"name": "accuracy", "type": "metrics", "pos": [200, 208]}, {"name": "SimRank", "type": "metrics", "pos": [233, 240]}]}, {"sentence": "The science fiction drama Sense8 debuted in June 2015 , which was written and produced by The Wachowskis and J. Michael Straczynski .", "entities": [{"name": "science fiction drama", "type": "else", "pos": [4, 25]}, {"name": "Sense8", "type": "else", "pos": [26, 32]}, {"name": "The Wachowskis", "type": "person", "pos": [90, 104]}, {"name": "J. Michael Straczynski", "type": "person", "pos": [109, 131]}]}, {"sentence": "While Eurotra never delivered a working MT system , the project made a far-reaching long-term impact on the nascent language industries in European member states , in particular among the southern countries of Greece , Italy , Spain , and Portugal .", "entities": [{"name": "Eurotra", "type": "else", "pos": [6, 13]}, {"name": "MT system", "type": "product", "pos": [40, 49]}, {"name": "European", "type": "else", "pos": [139, 147]}, {"name": "Greece", "type": "country", "pos": [210, 216]}, {"name": "Italy", "type": "country", "pos": [219, 224]}, {"name": "Spain", "type": "country", "pos": [227, 232]}, {"name": "Portugal", "type": "country", "pos": [239, 247]}]}, {"sentence": "Autoencoder has been successfully applied to the machine translation of human languages which is usually referred to as neural machine translation ( NMT ) .", "entities": [{"name": "Autoencoder", "type": "algorithm", "pos": [0, 11]}, {"name": "machine translation", "type": "task", "pos": [49, 68]}, {"name": "neural machine translation", "type": "task", "pos": [120, 146]}, {"name": "NMT", "type": "task", "pos": [149, 152]}]}, {"sentence": "Popular examples of fitness functions based on the probabilities include maximum likelihood estimation and hinge loss .", "entities": [{"name": "maximum likelihood estimation", "type": "metrics", "pos": [73, 102]}, {"name": "hinge loss", "type": "metrics", "pos": [107, 117]}]}, {"sentence": "Data mining is a related field of study , focusing on exploratory data analysis through unsupervised learning .", "entities": [{"name": "Data mining", "type": "field", "pos": [0, 11]}, {"name": "exploratory data analysis", "type": "task", "pos": [54, 79]}, {"name": "unsupervised learning", "type": "field", "pos": [88, 109]}]}, {"sentence": "Collaborative filtering encompasses techniques for matching people with similar interests and making recommender system on this basis .", "entities": [{"name": "Collaborative filtering", "type": "algorithm", "pos": [0, 23]}, {"name": "recommender system", "type": "product", "pos": [101, 119]}]}, {"sentence": "A number of WordNet-based word similarity algorithms are implemented in a Perl package called WordNet :: Similarity .", "entities": [{"name": "WordNet-based word similarity algorithms", "type": "algorithm", "pos": [12, 52]}, {"name": "Perl", "type": "program language", "pos": [74, 78]}, {"name": "WordNet :: Similarity", "type": "product", "pos": [94, 115]}]}, {"sentence": "Another paper , presented at the CVPR ( CVPR ) 2000 by Erik Miller , Nicholas Matsakis , and Paul Viola will also be discussed .", "entities": [{"name": "CVPR", "type": "conference", "pos": [33, 37]}, {"name": "CVPR", "type": "conference", "pos": [40, 44]}, {"name": "Erik Miller", "type": "researcher", "pos": [55, 66]}, {"name": "Nicholas Matsakis", "type": "researcher", "pos": [69, 86]}, {"name": "Paul Viola", "type": "researcher", "pos": [93, 103]}]}, {"sentence": "QC has not been evaluated against traditional modern clustering algorithms , aside from Jaccard index .", "entities": [{"name": "QC", "type": "algorithm", "pos": [0, 2]}, {"name": "clustering algorithms", "type": "else", "pos": [53, 74]}, {"name": "Jaccard index", "type": "metrics", "pos": [88, 101]}]}, {"sentence": "During the VEX Robotics World Championship , a Parade of Nations is held in Freedom Hall that includes hundreds of students from more than 30 countries .", "entities": [{"name": "VEX Robotics World Championship", "type": "else", "pos": [11, 42]}, {"name": "Parade of Nations", "type": "else", "pos": [47, 64]}, {"name": "Freedom Hall", "type": "location", "pos": [76, 88]}]}, {"sentence": "Other measures of accuracy include Single Word Error Rate ( SWER ) and Command Success Rate ( CSR ) .", "entities": [{"name": "Single Word Error Rate", "type": "metrics", "pos": [35, 57]}, {"name": "SWER", "type": "metrics", "pos": [60, 64]}, {"name": "Command Success Rate", "type": "metrics", "pos": [71, 91]}, {"name": "CSR", "type": "metrics", "pos": [94, 97]}]}, {"sentence": "They presented their method and results in SIGGRAPH 2000 .", "entities": [{"name": "SIGGRAPH 2000", "type": "conference", "pos": [43, 56]}]}, {"sentence": "The KDD Conference grew from KDD ( Knowledge Discovery and Data Mining ) workshops at AAAI conferences , which were started by Gregory I. Piatetsky-Shapiro in 1989 , 1991 , and 1993 , and Usama Fayyad in 1994 . Machinery | ACM .", "entities": [{"name": "KDD Conference", "type": "conference", "pos": [4, 18]}, {"name": "KDD", "type": "conference", "pos": [29, 32]}, {"name": "Knowledge Discovery and Data Mining", "type": "conference", "pos": [35, 70]}, {"name": "AAAI conferences", "type": "conference", "pos": [86, 102]}, {"name": "Gregory I. Piatetsky-Shapiro", "type": "researcher", "pos": [127, 155]}, {"name": "Usama Fayyad", "type": "researcher", "pos": [188, 200]}, {"name": "Machinery | ACM", "type": "conference", "pos": [211, 226]}]}, {"sentence": "He has been elected a Fellow of the Association for Computing Machinery ( ACM ) , the Institute of Electrical and Electronics Engineers ( IEEE ) , the International Association for Pattern Recognition ( IAPR ) , the Association for the Advancement of Artificial Intelligence ( AAAI ) , American Association for Advancement of Science ( AAAS ) , and the Society for Optics and Photonics Technology ( SPIE ) .", "entities": [{"name": "Association for Computing Machinery", "type": "conference", "pos": [36, 71]}, {"name": "ACM", "type": "conference", "pos": [74, 77]}, {"name": "Institute of Electrical and Electronics Engineers", "type": "organization", "pos": [86, 135]}, {"name": "IEEE", "type": "organization", "pos": [138, 142]}, {"name": "International Association for Pattern Recognition", "type": "conference", "pos": [151, 200]}, {"name": "IAPR", "type": "conference", "pos": [203, 207]}, {"name": "Association for the Advancement of Artificial Intelligence", "type": "conference", "pos": [216, 274]}, {"name": "AAAI", "type": "conference", "pos": [277, 281]}, {"name": "American Association for Advancement of Science", "type": "conference", "pos": [286, 333]}, {"name": "AAAS", "type": "conference", "pos": [336, 340]}, {"name": "Society for Optics and Photonics Technology", "type": "conference", "pos": [353, 396]}, {"name": "SPIE", "type": "conference", "pos": [399, 403]}]}, {"sentence": "Machine learning and data mining often employ the same methods and overlap significantly , but while machine learning focuses on prediction , based on known properties learned from the training data , data mining focuses on the discovery of ( previously ) unknown properties in the data ( this is the analysis step of knowledge discovery in databases ) .", "entities": [{"name": "Machine learning", "type": "field", "pos": [0, 16]}, {"name": "data mining", "type": "field", "pos": [21, 32]}, {"name": "machine learning", "type": "field", "pos": [101, 117]}, {"name": "data mining", "type": "field", "pos": [201, 212]}, {"name": "knowledge discovery in databases", "type": "task", "pos": [318, 350]}]}, {"sentence": "Indy is written in Java and therefore runs on most modern operating system s .", "entities": [{"name": "Indy", "type": "product", "pos": [0, 4]}, {"name": "Java", "type": "program language", "pos": [19, 23]}, {"name": "operating system", "type": "else", "pos": [58, 74]}]}, {"sentence": "NMF is an instance of nonnegative quadratic programming ( NQP ) , just like the support vector machine ( SVM ) .", "entities": [{"name": "NMF", "type": "algorithm", "pos": [0, 3]}, {"name": "nonnegative quadratic programming", "type": "algorithm", "pos": [22, 55]}, {"name": "NQP", "type": "algorithm", "pos": [58, 61]}, {"name": "support vector machine", "type": "algorithm", "pos": [80, 102]}, {"name": "SVM", "type": "algorithm", "pos": [105, 108]}]}, {"sentence": "The method is based on estimating the conditional probabilities using the nonparametric maximum likelihood method which leads", "entities": [{"name": "conditional probabilities", "type": "else", "pos": [38, 63]}, {"name": "nonparametric maximum likelihood", "type": "metrics", "pos": [74, 106]}]}, {"sentence": "The basic concepts involved in spectral estimation include autocorrelation , multi-D Fourier transform , mean square error and entropy .", "entities": [{"name": "autocorrelation", "type": "algorithm", "pos": [59, 74]}, {"name": "multi-D Fourier transform", "type": "algorithm", "pos": [77, 102]}, {"name": "mean square error", "type": "metrics", "pos": [105, 122]}, {"name": "entropy", "type": "metrics", "pos": [127, 134]}]}, {"sentence": "Application areas of kernel methods are diverse and include geostatistics , kriging , inverse distance weighting , 3D reconstruction , bioinformatics , chemoinformatics , information extraction and handwriting recognition .", "entities": [{"name": "kernel methods", "type": "algorithm", "pos": [21, 35]}, {"name": "geostatistics", "type": "field", "pos": [60, 73]}, {"name": "kriging", "type": "algorithm", "pos": [76, 83]}, {"name": "inverse distance weighting", "type": "algorithm", "pos": [86, 112]}, {"name": "3D reconstruction", "type": "task", "pos": [115, 132]}, {"name": "bioinformatics", "type": "field", "pos": [135, 149]}, {"name": "chemoinformatics", "type": "field", "pos": [152, 168]}, {"name": "information extraction", "type": "task", "pos": [171, 193]}, {"name": "handwriting recognition", "type": "task", "pos": [198, 221]}]}, {"sentence": "Robots can be autonomous or semi-autonomous and range from humanoids such as Honda ' s Advanced Step in Innovative Mobility ( ASIMO ) and TOSY ' s TOSY Ping Pong Playing Robot ( TOPIO ) to industrial robot s , medical operating robot s , patient assist robots , dog therapy robots , collectively programmed swarm robots , UAV drones such as General Atomics MQ-1 Predator , and even microscopic nano robots .", "entities": [{"name": "Honda", "type": "organization", "pos": [77, 82]}, {"name": "Advanced Step in Innovative Mobility", "type": "product", "pos": [87, 123]}, {"name": "ASIMO", "type": "product", "pos": [126, 131]}, {"name": "TOSY", "type": "organization", "pos": [138, 142]}, {"name": "TOSY Ping Pong Playing Robot", "type": "product", "pos": [147, 175]}, {"name": "TOPIO", "type": "product", "pos": [178, 183]}, {"name": "industrial robot", "type": "product", "pos": [189, 205]}, {"name": "medical operating robot", "type": "product", "pos": [210, 233]}, {"name": "patient assist robots", "type": "product", "pos": [238, 259]}, {"name": "dog therapy robots", "type": "product", "pos": [262, 280]}, {"name": "swarm robots", "type": "product", "pos": [307, 319]}, {"name": "UAV drones", "type": "product", "pos": [322, 332]}, {"name": "General Atomics MQ-1 Predator", "type": "product", "pos": [341, 370]}, {"name": "microscopic nano robots", "type": "product", "pos": [382, 405]}]}, {"sentence": "Freddy and Freddy II were robots built at the University of Edinburgh School of Informatics by Pat Ambler , Robin Popplestone , Austin Tate , and Donald Mitchie , and were capable of assembling wooden blocks in a period of several hours .", "entities": [{"name": "Freddy", "type": "product", "pos": [0, 6]}, {"name": "Freddy II", "type": "product", "pos": [11, 20]}, {"name": "University of Edinburgh School of Informatics", "type": "university", "pos": [46, 91]}, {"name": "Pat Ambler", "type": "researcher", "pos": [95, 105]}, {"name": "Robin Popplestone", "type": "researcher", "pos": [108, 125]}, {"name": "Austin Tate", "type": "researcher", "pos": [128, 139]}, {"name": "Donald Mitchie", "type": "researcher", "pos": [146, 160]}]}, {"sentence": "He spent his childhood years in Paris , France , where his parents had emigrated from Lithuania in the early 1920s .", "entities": [{"name": "Paris", "type": "location", "pos": [32, 37]}, {"name": "France", "type": "country", "pos": [40, 46]}, {"name": "Lithuania", "type": "country", "pos": [86, 95]}]}, {"sentence": "Previously , Dr. Paulos held the Cooper-Siegel Associate Professor Chair in the School of Computer Science at Carnegie Mellon University where he was faculty within the Human-Computer Interaction Institute with courtesy faculty appointments in the Robotics Institute and in the Entertainment Technology Center .", "entities": [{"name": "Dr. Paulos", "type": "researcher", "pos": [13, 23]}, {"name": "Cooper-Siegel Associate Professor Chair", "type": "else", "pos": [33, 72]}, {"name": "School of Computer Science", "type": "organization", "pos": [80, 106]}, {"name": "Carnegie Mellon University", "type": "university", "pos": [110, 136]}, {"name": "Human-Computer Interaction Institute", "type": "university", "pos": [169, 205]}, {"name": "Robotics Institute", "type": "university", "pos": [248, 266]}, {"name": "Entertainment Technology Center", "type": "university", "pos": [278, 309]}]}, {"sentence": "In 1969 Victor Scheinman at Stanford University invented the Stanford arm , an all-electric , 6-axis articulated robot designed to permit an arm solution .", "entities": [{"name": "Victor Scheinman", "type": "researcher", "pos": [8, 24]}, {"name": "Stanford University", "type": "university", "pos": [28, 47]}, {"name": "Stanford arm", "type": "product", "pos": [61, 73]}, {"name": "6-axis articulated robot", "type": "product", "pos": [94, 118]}, {"name": "arm solution", "type": "else", "pos": [141, 153]}]}, {"sentence": "The creation and implementation of chatbots is still a developing area , heavily related to artificial intelligence and machine learning , so the provided solutions , while possessing obvious advantages , have some important limitations in terms of functionalities and use cases .", "entities": [{"name": "chatbots", "type": "product", "pos": [35, 43]}, {"name": "artificial intelligence", "type": "field", "pos": [92, 115]}, {"name": "machine learning", "type": "field", "pos": [120, 136]}]}, {"sentence": "In terms of freely available resources , Carnegie Mellon University ' s Sphinx toolkit is one place to start to both learn about speech recognition and to start experimenting .", "entities": [{"name": "Carnegie Mellon University", "type": "university", "pos": [41, 67]}, {"name": "Sphinx toolkit", "type": "product", "pos": [72, 86]}, {"name": "speech recognition", "type": "task", "pos": [129, 147]}]}, {"sentence": "The formal RoboCup competition was preceded by the ( often unacknowledged ) first International Micro Robot World Cup Soccer Tournament ( MIROSOT ) held by KAIST in Taejon , Korea , in November 1996 .", "entities": [{"name": "RoboCup competition", "type": "else", "pos": [11, 30]}, {"name": "International Micro Robot World Cup Soccer Tournament", "type": "else", "pos": [82, 135]}, {"name": "MIROSOT", "type": "else", "pos": [138, 145]}, {"name": "KAIST", "type": "university", "pos": [156, 161]}, {"name": "Taejon", "type": "location", "pos": [165, 171]}, {"name": "Korea", "type": "country", "pos": [174, 179]}]}, {"sentence": "In addition to the standard hinge loss math ( 1-yf ( x ) ) _ + / math for labeled data , a loss function math ( -1 | f ( x ) | ) _ + / math is introduced over the unlabeled data by letting mathy = \\ operatorname { sign } { f ( x ) } / math .", "entities": [{"name": "hinge loss", "type": "metrics", "pos": [28, 38]}, {"name": "loss function", "type": "else", "pos": [91, 104]}]}, {"sentence": "In particular , RLS is designed to minimize the mean squared error between the predicted values and the TRUE labels , subject to regularization .", "entities": [{"name": "RLS", "type": "else", "pos": [16, 19]}, {"name": "mean squared error", "type": "metrics", "pos": [48, 66]}]}, {"sentence": "Essentially , this combines maximum likelihood estimation with a regularization procedure that favors simpler models over more complex models .", "entities": [{"name": "maximum likelihood estimation", "type": "algorithm", "pos": [28, 57]}, {"name": "regularization procedure", "type": "algorithm", "pos": [65, 89]}]}, {"sentence": "The true-positive rate is also known as sensitivity , recall or probability of detection math to the discrimination threshold ) of the detection probability in the y-axis versus the cumulative distribution function of the false-alarm probability on the x-axis .", "entities": [{"name": "true-positive rate", "type": "metrics", "pos": [4, 22]}, {"name": "sensitivity", "type": "metrics", "pos": [40, 51]}, {"name": "recall", "type": "metrics", "pos": [54, 60]}, {"name": "probability of detection", "type": "metrics", "pos": [64, 88]}, {"name": "discrimination threshold", "type": "else", "pos": [101, 125]}, {"name": "cumulative distribution function", "type": "algorithm", "pos": [182, 214]}, {"name": "false-alarm probability", "type": "metrics", "pos": [222, 245]}]}, {"sentence": "In English , WordNet is an example of a semantic network .", "entities": [{"name": "English", "type": "else", "pos": [3, 10]}, {"name": "WordNet", "type": "product", "pos": [13, 20]}]}, {"sentence": "Prolonged use of speech recognition software in conjunction with word processor s has shown benefits to short-term-memory restrengthening in brain AVM patients who have been treated with resection .", "entities": [{"name": "speech recognition software", "type": "product", "pos": [17, 44]}, {"name": "word processor", "type": "product", "pos": [65, 79]}, {"name": "brain AVM", "type": "else", "pos": [141, 150]}]}, {"sentence": "Its founding editor-in-chiefs were Ron Sun , Vasant Honavar , and Gregg Oden ( from 1999 to 2014 ) .", "entities": [{"name": "Ron Sun", "type": "researcher", "pos": [35, 42]}, {"name": "Vasant Honavar", "type": "researcher", "pos": [45, 59]}, {"name": "Gregg Oden", "type": "researcher", "pos": [66, 76]}]}, {"sentence": "Their ' parallel ' distinction , as opposed to a serial manipulator , is that the end effector ( or ' hand ' ) of this linkage ( or ' arm ' ) is directly connected to its base by a number of ( usually three or six ) separate and independent linkages working simultaneously .", "entities": [{"name": "serial manipulator", "type": "product", "pos": [49, 67]}, {"name": "end effector", "type": "else", "pos": [82, 94]}]}, {"sentence": "His thesis advisor was Professor Cordell Green , and his thesis / oral committee included Professors Edward Feigenbaum Joshua Lederberg , Paul Cohen , Allen Newell , Herbert Simon , .", "entities": [{"name": "Cordell Green", "type": "researcher", "pos": [33, 46]}, {"name": "Edward Feigenbaum", "type": "researcher", "pos": [101, 118]}, {"name": "Joshua Lederberg", "type": "researcher", "pos": [119, 135]}, {"name": "Paul Cohen", "type": "researcher", "pos": [138, 148]}, {"name": "Allen Newell", "type": "researcher", "pos": [151, 163]}, {"name": "Herbert Simon", "type": "researcher", "pos": [166, 179]}]}, {"sentence": "Such functions include the mean squared error , root mean squared error , mean absolute error , relative squared error , root relative squared error , relative absolute error , and others .", "entities": [{"name": "mean squared error", "type": "metrics", "pos": [27, 45]}, {"name": "root mean squared error", "type": "metrics", "pos": [48, 71]}, {"name": "mean absolute error", "type": "metrics", "pos": [74, 93]}, {"name": "relative squared error", "type": "metrics", "pos": [96, 118]}, {"name": "root relative squared error", "type": "metrics", "pos": [121, 148]}, {"name": "relative absolute error", "type": "metrics", "pos": [151, 174]}]}, {"sentence": "There are bindings in Python , Java and MATLAB / OCTAVE .", "entities": [{"name": "Python", "type": "program language", "pos": [22, 28]}, {"name": "Java", "type": "program language", "pos": [31, 35]}, {"name": "MATLAB", "type": "product", "pos": [40, 46]}, {"name": "OCTAVE", "type": "program language", "pos": [49, 55]}]}, {"sentence": "An implementation in MATLAB can be found on the .", "entities": [{"name": "MATLAB", "type": "product", "pos": [21, 27]}]}, {"sentence": "John McCarthy is one of the founding fathers of artificial intelligence , together with Alan Turing , Marvin Minsky , Allen Newell , and Herbert A. Simon .", "entities": [{"name": "John McCarthy", "type": "researcher", "pos": [0, 13]}, {"name": "artificial intelligence", "type": "field", "pos": [48, 71]}, {"name": "Alan Turing", "type": "researcher", "pos": [88, 99]}, {"name": "Marvin Minsky", "type": "researcher", "pos": [102, 115]}, {"name": "Allen Newell", "type": "researcher", "pos": [118, 130]}, {"name": "Herbert A. Simon", "type": "researcher", "pos": [137, 153]}]}, {"sentence": "A parallel manipulator is a mechanical system that uses several serial manipulator s to support a single platform , or end-effector .", "entities": [{"name": "serial manipulator", "type": "product", "pos": [64, 82]}, {"name": "end-effector", "type": "else", "pos": [119, 131]}]}, {"sentence": "GATE includes an information extraction system called ANNIE ( A Nearly-New Information Extraction System ) which is a set of modules comprising a tokenizer , a gazetteer , a sentence splitter , a Part-of-speech tagging , a Named entity recognition transducer and a coreference tagger .", "entities": [{"name": "GATE", "type": "product", "pos": [0, 4]}, {"name": "information extraction", "type": "task", "pos": [17, 39]}, {"name": "ANNIE", "type": "product", "pos": [54, 59]}, {"name": "A Nearly-New Information Extraction System", "type": "product", "pos": [62, 104]}, {"name": "tokenizer", "type": "else", "pos": [146, 155]}, {"name": "gazetteer", "type": "else", "pos": [160, 169]}, {"name": "sentence splitter", "type": "else", "pos": [174, 191]}, {"name": "Part-of-speech tagging", "type": "task", "pos": [196, 218]}, {"name": "Named entity recognition transducer", "type": "product", "pos": [223, 258]}, {"name": "coreference tagger", "type": "product", "pos": [265, 283]}]}, {"sentence": "He graduated from Moscow State University and in November 1978 , he left for the United States thanks to the personal intervention of Senator Edward M. Kennedy ..", "entities": [{"name": "Moscow State University", "type": "university", "pos": [18, 41]}, {"name": "United States", "type": "country", "pos": [81, 94]}, {"name": "Senator Edward M. Kennedy", "type": "person", "pos": [134, 159]}]}, {"sentence": "In 2017 , the DeepMind AlphaGo team received the inaugural IJCAI Marvin Minsky medal for Outstanding Achievements in AI .", "entities": [{"name": "DeepMind AlphaGo team", "type": "organization", "pos": [14, 35]}, {"name": "inaugural IJCAI Marvin Minsky medal", "type": "else", "pos": [49, 84]}, {"name": "AI", "type": "field", "pos": [117, 119]}]}, {"sentence": "Other ways anomalous propagation is recorded is by troposcatter s causing irregularities in the troposphere , scattering due to meteor s , refraction in the ionized regions and layers of the ionosphere , and reflection from the ionosphere .", "entities": [{"name": "anomalous propagation", "type": "else", "pos": [11, 32]}, {"name": "troposcatter", "type": "else", "pos": [51, 63]}, {"name": "troposphere", "type": "else", "pos": [96, 107]}, {"name": "ionized regions", "type": "else", "pos": [157, 172]}, {"name": "ionosphere", "type": "else", "pos": [191, 201]}, {"name": "ionosphere", "type": "else", "pos": [228, 238]}]}, {"sentence": "Natural language processing ( NLP ) is a subfield of linguistics , computer science , information engineering , and artificial intelligence concerned with the interactions between computers and human ( natural ) languages , in particular how to program computers to process and analyze large amounts of natural language data .", "entities": [{"name": "Natural language processing", "type": "field", "pos": [0, 27]}, {"name": "NLP", "type": "field", "pos": [30, 33]}, {"name": "linguistics", "type": "field", "pos": [53, 64]}, {"name": "computer science", "type": "field", "pos": [67, 83]}, {"name": "information engineering", "type": "field", "pos": [86, 109]}, {"name": "artificial intelligence", "type": "field", "pos": [116, 139]}]}, {"sentence": "Other active youth-led climate groups include Extinction Rebellion , the Sunrise Movement , SustainUS , the , among others working at both the transnational and local levels .", "entities": [{"name": "Extinction Rebellion", "type": "organization", "pos": [46, 66]}, {"name": "Sunrise Movement", "type": "organization", "pos": [73, 89]}, {"name": "SustainUS", "type": "organization", "pos": [92, 101]}]}] |