text stringlengths 4 222k | label int64 0 4 |
|---|---|
Automatically generating new data is a critical component within modern machine learning pipelines. During training, data augmentation can expose models to a larger portion of potential input space, consistently leading to better generalization and performance (Simard et al., 1998; Krizhevsky et al., 2012; Perez and Wa... | 0 |
Estimation of hidden Marknv model (HMM) is usually obtained by the method of maximum likelihood (ML) [1, 10, 6] assuming that the size of the training data is large enough to provide robust estimates. This paper investigates maximum a posteriori (MAP) estimate of continuous density hidden Markov models (CDHMM). The MAP... | 0 |
Différentes techniques issues du traitement automatique des langues (TAL) ont été utilisées au fil des ans pour traiter la tâche de classification de textes. Avec l'émergence de corpus de taille plus importante et l'avènement de l'apprentissage profond, les architectures de réseaux neuronaux sont devenues de plus en pl... | 0 |
The SemEval-2015 shared task on "Sentiment Analysis in Twitter" (Rosenthal et al., 2015) is a rerun of the shared task from SemEval-2014 (Rosenthal et al., 2014) with three new subtasks. While subtasks A and B were identical to the tasks of SemEval-2014 and dealt with the identification of polarity in a given message, ... | 0 |
Emotion recognition in conversation (ERC) is a task within the sphere of emotion recognition. ERC aims to predict the emotion of each utterance in a conversation. With the recent advances of dialogue research, ERC has gained popularity due to its potential to support downstream applications such as affective dialog sys... | 0 |
Natural language processing has developed in the last years in several directions that often go far beyond the original goal of mapping natural laua-~uage expressions into formal internal representazions. ?roblems concerned with discourse modeling, reasoning about beliefs, knowledge and wants of speaker and hearer, exp... | 0 |
The most important problem in all speech recognition systems is the inherent uncertainty associated with the acoustic-phonetic decoding process at the basis of such a system. One approach taken in many existing system to overcome these difficulties is to integrate higher level knowledge sources that have a certain a-pr... | 0 |
Computers have been powerful support tools for various kinds of human activities. In particular, high performance personal workstations provide convenient and friendly environments for office workers and engineers. Conventional systems, however, provide only basic support tools such as text editors, text formatters, sp... | 0 |
posed the problem of "miniature" language acquisition based on <sentence, image> pairs as a "touchstone" for cognitive science. In this task, an artificial system is confronted with a reduced version of the problem of language acquisition faced by the child, that involves both the extraction of meaning from the image, ... | 0 |
Recognition of named entities in natural language text is an important subtask of information extraction and thus bears importance for modern text mining and information retrieval applications. The need to identify named entities such as persons, locations, organizations and places, arises both in applications where th... | 0 |
Conversational bots are increasingly popular among the general population which is correlated with an increase in bot abuse (Cercas Curry and Rieser, 2018) . Analysis of the chat logs of an Alexa Prize 1 competition social bot shows that more than 10% of the conversations contain some level of offensiveness. Recently, ... | 0 |
Building supervised learning models is like operating a manufacturing plant. Raw materials(data) need to be refined and processed(annotated) as a precursor to final assembly. Some manufacturing plants rely on a supply chain (outsource annotation) while others are vertically integrated (annotate in house). According to ... | 0 |
Cross-Lingual Word Embedding (CLWE) techniques have recently received significant attention as an effective means to support Natural Language Processing applications for low-resource languages, e.g., machine translation (Artetxe et al., 2018b) and transfer learning (Peng et al., 2021) .The most successful CLWE models a... | 0 |
The ideas of text in-filling coupled with selfsupervised pre-training of deep Transformer networks on large text corpora have dramatically changed the landscape in Natural Language Understanding. BERT (Devlin et al., 2019) and its successive refinements RoBERTa , ALBERT (Lan et al., 2019) implement this recipe and have... | 0 |
When we say that a parser is robust it is not quite clear what we mean, since the notion of robustness reflects in fact an informal collection of aspects related to the improper use or the exceptional behavior of the parser. One aspect that is mentioned frequently in this context, concerns the adequate behavior of the ... | 0 |
We are building a system that offers users expert advice (on health, but the particular domain is unimportant) by means of multiple-turn natural language dialogue. The work presented in this paper concerns largely the generation of the meanings of utterances, meanings which contain all the information necessary for the... | 0 |
Reading Comprehension (RC) is the task of reading a body of text and answering questions about it. It requires a deep understanding of the information presented in order to reason about entities, actions, events, and their interrelationships. This necessitates language understanding skills as well as the cognitive abil... | 0 |
In current phrase-based statistical machine translation systems such as Moses 1 (Koehn et al., 2007) , the translation model is defined in terms of phrase pairs (biphrases) extracted from a bilingual corpus as follows. The corpus is first word-aligned using a word alignment heuristic (Och and Ney, 2003) . The phrase ex... | 0 |
Harnessing the full potential in text data has always been a key task for the NLP and ML communities. The properties hidden under the inherent high dimensionality of text are of major importance in tasks such as text categorization and opinion mining.Although simple models like bag-of-words manage to perform well, the ... | 0 |
In this work we are concerned with building statistical models for parse disambiguation -choosing a correct analysis out of the possible analyses for a sentence. Many machine learning algorithms for classification and ranking require data to be represented as real-valued vectors of fixed dimensionality. Natural languag... | 0 |
One of the natural goals for research on speech signals is recognizing emotions of humans [Chen 1987; Oppenheim 1976; Cowie 2001] ; it has gained growing amounts of interest over the last 20 years. A study conducted by Shirasawa et al. showed that SER could be made by ICA and attain an 87% average recognition ratio [Sh... | 0 |
In text generation systems (as summarizers, question-answering systems, etc.), coherence is an essential characteristic in order to produce comprehensible texts. As such, studies and theories on coherence ( (Mann and Thompson, 1998) , (Grosz et al., 1995) ) have supported applications that involve text generation ( (Se... | 0 |
Mental illnesses, such as depression and post traumatic stress disorder (PTSD) represent a large share of the global burden of disease (Üstün et al., 2004; Mathers and Loncar, 2006) , but are underdiagnosed and undertreated around the world (Prince et al., 2007) . Previous research has demonstrated the important role o... | 0 |
Background. Vossian Antonomasia (VA) is a stylistic device which attributes a certain property to a person by naming another (more well-known, more popular) person as a reference point. For instance, when Jim Koch is described as "the Steve Jobs of Beer" (Fallows, 2014) , certain qualities of Steve Jobs, be it entrepre... | 0 |
Twitter and other social media services have received considerable attentions in recent years. Users provide hundreds of millions microblogs through them everyday. The informative data has been relied on by many applications, such as sentiment analysis (Jiang et al., 2011; Meng et al., 2012) , event detection (Sakaki e... | 0 |
Entrainment, also called accommodation or alignment, is the tendency of human interlocutors to adapt their behavior to each other to become more similar. This affects many linguistic features such as referring expressions (Brennan and Clark, 1996) , phonetics (Pardo, 2006) , syntax (Reitter et al., 2006) , linguistic s... | 0 |
In Australia bushfire management is a state and territory government responsibility and each jurisdiction has its own agency which takes the lead in coordinating community preparedness and responding to bushfires when they occur. For example, the Rural Fire Services (RFS) in NSW, the Country Fire Authority (CFA) in Vic... | 0 |
In this paper, we introduce a novel method for mapping high-level instructions to commands in an external environment. These instructions specify goals to be achieved without explicitly stating all the required steps. For example, consider the first instruction in Figure 1 -"open control panel." The three GUI commands ... | 0 |
Recently, automatically generating image descriptions has attracted considerable interest in the fields of computer vision and nature language processing. Such a task is easy to humans but highly non-trivial for machines as it requires not only capturing the semantic information from images (e.g., objects and actions) ... | 0 |
Relation classification is an important Natural Language Processing (NLP) task which is normally used as an intermediate step in many complex NLP applications such as question-answering and automatic knowledge base construction. Since the last decade there has been increasing interest in applying machine learning appro... | 0 |
Hindi is an official language of India and is written in Devanagari script. Urdu is the national language of Pakistan, and also one of the state languages in India, and is written in Perso-Arabic script. Hindi inherits its vocabulary from Sanskrit while Urdu descends from several languages including Arabic, Farsi (Pers... | 0 |
Spelling errors are common in our daily life, caused typically by human writing, automatic speech recognition, and optical character recognition systems. Among these errors, misspelling a character frequently occurs due to the similarity between characters. In Chinese, many characters are phonologically and visually si... | 0 |
Resolving anaphoric relations automatically requires annotated data for training and testing. Anaphora and coreference resolution systems have been tested and evaluated on different genres, mainly news articles and dialogue. However, for scientific text, annotated data are scarce and coreference resolution systems are ... | 0 |
The performance of a speech recognition system is often severely degraded when there is a mismatch between the acoustic conditions of the training and the application environments. This mismatch may come from various sources, such as additive noise, channel distortion, different speaker characteristics, and different s... | 0 |
As large amounts of language data have become available, approaches to sentence-level processing tasks such as parsing, language modeling, named-entity detection and machine translation have become increasingly data-driven and empirical. Models for these tasks can be trained to capture the distributions of phenomena in... | 0 |
For any unsegmented language, especially East Asian languages such as Chinese, Japanese and Thai, word segmentation is almost an inevitable first step in natural language processing. In fact, it is becoming increasingly important lately because of the growing interest in processing user-generated media, such as Twitter... | 0 |
Answering emails sent to a company by its customers -to take just one example among many similar text-processing tasks -requires a reliable understanding of the content of incoming messages. This understanding can currently only be done by humans, and represents the main bottleneck to a complete automation of the proce... | 0 |
Human-annotated corpora are essential for the development and evaluation of natural language processing methods. However, creating such corpora is expensive and time-consuming. While remote annotation processes on the web such as crowdsourcing provide a way to obtain numerous annotations in short time, the bottleneck o... | 0 |
Open-domain dialog systems have typically been modeled using end-to-end approaches, more specifically encoder-decoder architectures (Sordoni et al., 2015; Serban et al., 2017 Serban et al., , 2016 Vinyals and Le, 2015) . These seq2seq models are commonly trained on a maximum likelihood objective, which leads to repetit... | 0 |
People engage in argumentation in various contexts, both online and in the real life. Existing definitions of argumentation do not solely focus on giving reasons and laying out a logical framework of premises and conclusions, but also highlight its social purpose which is to convince or to persuade (O'Keefe, 2011; van ... | 0 |
In recent years, contrastive learning (CL) has been widely applied to self-supervised representation learning and led to major advances across computer vision (CV) Chen et al., 2020b) , speech (Saeed et al., 2021; , and natural language processing (NLP) (Fang and Xie, 2020; Gao et al., 2021; Yan et al., 2021) . The bas... | 0 |
Knowledge bases (KBs) such as Freebase (Bollacker et al., 2008) and DBpedia (Auer et al., 2007) are fundamental resources for many intelligent applications. Currently, the construction and updating of KBs, directly or indirectly, rely on human labor. Keeping KBs up-to-date by humans is cost intensive and impractical. T... | 0 |
Topic models provide a high-level view of the main themes of a document collection . Document collections, however, are often not in a single language, driving the development of multilingual topic models. These models discover topics that are consistent across languages, providing useful tools for multilingual text an... | 0 |
Using subword segmentation has become de-facto standard in Neural Machine Translation (Bojar et al., 2018; Barrault et al., 2019) . Byte Pair Encoding (BPE) (Sennrich et al., 2016) is the dominant approach to subword segmentation. It keeps the common words intact while splitting the rare and unknown ones into a sequenc... | 0 |
The aim of Natural Language Understanding (NLU) in task-oriented dialogue systems is to identify the user's need from their utterance . This comprises the following crucial information: 1) intents, what the user intends to do, and 2) (typically predefined) slots, associated arguments of the intent (Tur et al., 2010; Tu... | 0 |
Widespread data collection from news sources and microblogs has produced massive textual data streams that are challenging to process and analyze. The detection of emerging events from data streams such as Twitter has received growing attention from researchers. Many methods focus on detecting specific, "bursty" events... | 0 |
Semantic parsers translate natural language (NL) utterances into formal meaning representations. In particular, task-oriented semantic parsers map user-issued utterances (e.g. Find papers in ACL) into machine-executable programs (e.g. a database query), play a key role in providing natural language interfaces to applic... | 0 |
Typically, sentiment analysis is modeled as a three-class classification task, marking utterances as either positive, negative or neutral. In some cases, this may be accompanied with a degree of polarity. However, that still treats the task as a one-dimensional one along the scale of general polarity. In this paper, we... | 0 |
Visual Question Answering (VQA) (Antol et al., 2015) , the task of answering questions about visual content, was proposed to facilitate the development of models with human-like visual and linguistic understanding. However, existing VQA models often exploit superficial statistical biases to produce responses, instead o... | 0 |
A central function of a spoken dialog system (SDS) is to estimate the user's intention based on the utterances. The information gathered across multiple turns needs to be combined and understood in context after automatic speech recognition (ASR). Traditionally, this has been addressed by dialog models and data structu... | 0 |
The GREC task is about how to generate appropriate references to an entity in the context of a piece of discourse longer than a sentence. Rather than requiring participants to generate referring expressions from scratch, the GREC data provides sets of possible referring expressions for selection. As this is a new refer... | 0 |
The medical concept normalisation task aims to assign a corresponding identifier from a standard terminology to text descriptions. Depending on the domain, descriptions may vary from formal medical jargon terms (e.g. "Dizziness") to more informal and colloquial expressions that rather explain how the patient feels (e.g... | 0 |
Among the various social media platforms, microblogging services such as Twitter 1 have become popular communication tools. This is due to the easy accessibility of microblogging platforms via internet or mobile phones, and due to the need for a fast mode of communication that microblogging satisfies: Twitter messages ... | 0 |
Opinion mining, also called sentiment analysis, is the task of extracting and analyzing opinions, sentiments, evaluations or feelings from user-generated contents such as reviews, discussion groups and blogs. Due to its wide range of applications such as analysis of customer reviews (Hu & Liu, 2004) and reputation mana... | 0 |
Chunking is recognized as series of processesfirst identifying proper chunks from a sequence of tokens (such as words), and second classifying these chunks into some grammatical classes. Various NLP tasks can be seen as a chunking task. Examples include English base noun phrase identification (base NP chunking), Englis... | 0 |
Modern statistical machine translation systems are becoming more accurate, but also more complex. To cope with increased system complexity, it is convenient to carve systems into modules that can be separately developed, improved, and tested. In this paper, we explore the cost of such modularization on overall system p... | 0 |
What is the most difficult example in the Stanford Natural Language Inference (SNLI) data set (Bowman et al., 2015) or in the Stanford Sentiment Treebank (SSTB) (Socher et al., 2013) ? A priori the answer is not clear. How does one quantify the difficulty of an example and does it pertain to a specific model, or more g... | 0 |
Encoder-decoder networks (Pollack, 1990; Chrisman, 1991; Forcada andÑeco, 1997; Kalchbrenner and Blunsom, 2013) are the current prevailing architecture for neural machine translation (NMT). Various architectures have been used in the general framework of encoder and decoder networks such as recursive auto-encoders (Pol... | 0 |
In recent years, there has been a huge amount of research on applying self-attention models to NLP tasks. Transformer (Vaswani et al., 2017) is the most common architecture, which can capture long-range dependencies by using a self-attention mechanism over a set of vectors. To encode the sequential structure of sentenc... | 0 |
For most of the 7000 languages of the world, no NLP resources exist (Joshi et al., 2020; Mager et al., 2018) . As a response to this situation, more and more initiatives emerged in recent years that work on NLP applications for underrepresented and low-resource languages (Orife et al., 2020; Nekoto et al., 2020; Mager ... | 0 |
Text summarization has recently received increased attention with the rise of deep learning-based endto-end models, both for extractive and abstractive variants. However, so far, only single-document summarization has profited from this trend. Multidocument summarization (MDS) still suffers from a lack of established l... | 0 |
Recently, various cognitive problems, such as image (Deng et al., 2009) and speech recognition (Hinton et al., 2012) , have been solved by deep learning-based machine learning frameworks. A deep learning-based cognitive system needs a large number of datasets for model training to prevent the model from over-training (... | 0 |
Class based approach to lexical semantics such as presented in Levin (1993) provides a straightforward way of describing a large number of verbs in a compact and generalized way. The main assumption is the correlation between the syntactic behaviour of verbs as reflected in diathesis alternations and their semantic pro... | 0 |
The goal of sentence compression is to shorten sentences while preserving their grammaticality and important information. It has recently attracted much attention because of its wide range of applications, especially in summarization (Jing, 2000) and headline generation (which can be viewed as summarization with very s... | 0 |
Text to speech synthesis systems(TTS) help users to interact with computer through speech. System with speech interaction is advantageous for physically challenged, especially visually impaired. Important challenge in speech synthesis is generating synthesized speech which is both intelligible and natural.TTS systems s... | 0 |
A growing interest in interactive conversational agents and robots has motivated research focus on spoken language understanding (SLU). As an essential part of spoken dialog system (SDS), SLU analyzes user input, and provides the dialog system with information to make a response. In conversations, dialog act (DA) repre... | 0 |
Commonsense refers to the skill of making presumptions regarding the physical form, use, behaviour, interaction with other objects etc. that is derived from the naive physics as well as the humans' folk psychology that develops because of the frequent experience that we have as a result of our day to day interaction wi... | 0 |
In this paper, I would like to address a methodological question that starts from the following observation: There is a growing number of corpus studies in the area of the use of coherence relations and discourse markers in natural texts.Most of these studies try to establish a relationship between the coherence relati... | 0 |
Entity linking (EL) is the task of grounding mentions to a reference knowledge base (also referred to as knowledge graph). With a plethora of applications, including but not limited to information extraction (Hoffart et al., 2011) and automatic knowledge base construction (Gao et al., 2018) , EL is one of the most acti... | 0 |
In recent years, phrase-based translation models have become the mainstream of statistical machine translation, because they can represent context-based word selection and local word reordering better than word-based translation models. Previous phrased-based translation models [1, 2] , however, are not effective for g... | 0 |
A language model as a t)ost-processor is esse, ntial to a recognizer of speech or characters in order to determine the approi)riate word se, que, n(:e and henc.e the semantics of an inI)ut line of text or utterance. It is well known that an N-gram statistics language model is just as effective as, t)ut nmch more eifici... | 0 |
The combination of a preposition with a singular count noun, illustrated in (1) with the preposition unter, is a frequent construction in written and spoken German. From a theoretical perspective, constructions like (1) are interesting since they seem to violate the near universal rule that determiners should accompany... | 0 |
The Semantic Similarity is defined as a concept whereby a set of words are assigned a metric based on the likeliness of the semantic content. It is easy for humans with their cognitive abilities to judge the semantic similarity between two given words or concepts. For example, a human can quite easily say that the word... | 0 |
L'acquisition d'une langue seconde (L2), en particulier à l'âge adulte, est toujours marquée par des difficultés en termes d'articulation et de perception des contrastes phonémiques non-natifs, c'est-àdire des sons de la L2 qui n'existent pas ou ne sont pas phonologiquement distincts dans la langue maternelle (L1) des ... | 0 |
Neural networks have been extremely successful statistical models of text in language modeling and machine translation. Despite differences in model architectures, state of the art neural nets generate sequences from left to right (Vaswani et al., 2017; Jozefowicz et al., 2016; . Although in some sense humans produce a... | 0 |
Corpus-based approaches to machine translation (MT) have achieved much progress over the last decades. Despite a high performance on average, these approaches can often produce translations with severe errors. Input sentences featuring linguistic phenomena that are not sufficiently covered by the utilized models cannot... | 0 |
In recent years, the automatic knowledge extraction using Wikipedia has attracted significant research interest in research fields, such as the semantic web. As a valuable source of structured knowledge, Wikipedia infoboxes have been utilized to build linked open data (Suchanek et al., 2007; Bollacker et al., 2008; Biz... | 0 |
Coreference resolution is the task of identifying all mentions in a text that refer to the same real-world entity. However, it is time-consuming and expensive to collect the large amounts of data from expert annotators that are required to train high-performance coreference resolution systems. A rapid and costeffective... | 0 |
Text understanding is one of the fundamental tasks in Natural Language Processing areas. These years we have seen significant progress in applying neural networks to text analysis applications. Recurrent neural network is widely used because of its effective capability of capturing the sequential information. Long shor... | 0 |
In this paper we investigate how a traditional lexical syntactico-semantic resource, namely "Les Verbes Français" (The French Verbs, henceforth abbreviated LVF) (Dubois and Dubois-Charlier, 1997; François et al., 2007) , can best be converted into a standardised and normalised linked open data model. Our motivation is ... | 0 |
Si la multimodalité prend une part croissante dans les études linguistiques, les corpus annotés d'enregistrements vidéos, prenant en compte la gestualité co-verbale, restent encore assez peu nombreux à l'heure actuelle. Cela ne signifie pas cependant qu'une réflexion n'a pas été conduite sur l'annotation de ces phénomè... | 0 |
When working with natural language processing (NLP) for historical text, one problem is that there are often not large enough amounts of annotated corpus data available for training NLP tools specifically aimed at handling historical text. Nevertheless, using existing NLP tools as they are is rarely an option, since th... | 0 |
There are many tools that natural language processing (NLP) can offer disadvantaged readers to aid them in understanding a document. Readers may be at a disadvantage due to poor sight, to cognitive disabilities, or simply to reading in a language other than their native one (L1). This paper addresses that last case. Fo... | 0 |
This paper presents a study on German infinitive-verb compounds with an inflected head verb and an infinitive modifier. Specifically, we consider infinitive-verb compounds with the head verbs bleiben and lassen which allow for an idiomatic, i.e., conventionalized figurative interpretation in addition to a literal inter... | 0 |
Automatic Speech Recognition (ASR) systems are rapidly becoming part of our everyday lives through voice assistants such as Siri, Alexa, and Google Assistant. Since these voice assistants can now perform various day-to-day tasks exceedingly well, they have now become an integral part of many devices such as phones, tel... | 0 |
Users expect data-to-text NLG systems to generate textual summaries which are accurate. However, many neural NLG systems in particular generate texts which are factually incorrect.The most reliable way to assess the accuracy of a generated text is to ask human annotators to carefully fact-check the text. However this i... | 0 |
The work presented here is part of the PROTEUS (-PROtotype TExt Understanding System) system currently under development at the Courant Institute of Mathematical Sciences, New York University. l The objective of our research is to understand short natural language texts about equipment. Our texts at present are CASualt... | 0 |
Interpretability is a pressing concern for many current NLP models. As they become increasingly complex and learn decision-making functions from data, ensuring our ability to understand why a particular decision occurred is critical.Part of that development has been the incorporation of attention mechanisms (Bahdanau e... | 0 |
Text error detection techniques have been widely used in a variety of applications, e.g., spell checkers and Optical Character Recognition (OCR) readers. Studies have explored methods to improve the accuracy of word correction, with several techniques commonly used for solving the problem. Dictionary-matching, common e... | 0 |
Weighted Logic Programming (WLP) is a declarative approach to specifying and reasoning about dynamic programming algorithms and chart parsers. WLP is a generalization of bottom-up logic programming where proofs are assigned weights by combining the weights of the axioms used in the proof, and the weight of a theorem is... | 0 |
A document retrieval system typically solves a text ranking problem defined as follows: given a query x, a relevance score s(x, y) is computed for each document y in the target collection D. Thus, the text ranking problem can be equivalently cast as a relevance-based binary classification problem (Lin et al., 2020) , w... | 0 |
Neural approaches rely on labeled data sets for training. For many tasks and languages, such data is either scarce or not available at all. Knowledgebased weak supervision tackles this problem by employing labeling functions (LFs). LFs are manually specified properties, e.g. keywords, that trigger the automatic annotat... | 0 |
Research on generating textual descriptions of images has made great progress in recent years with the introduction of encoder-decoder architectures (Xu et al., 2015; Johnson et al., 2016; Venugopalan et al., 2017; Karpathy and Fei-Fei, 2017; Anderson et al., 2018; Lu et al., 2018b; Aneja et al., 2018) . Those models a... | 0 |
Semantic role labeling (SRL) is one of the fundamental tasks in natural language processing because of its important role in information extraction (Bastianelli et al., 2013) , statistical machine translation (Aziz et al., 2016; Xiong et al., 2012) , and so on.However, state-of-the-art performance of Chinese SRL is sti... | 0 |
Systematic relationships between pairs of strings are at the core of problems such as transliteration (Knight and Graehl, 1998) , morphology (Dreyer and Eisner, 2011) , cross-document coreference resolution (Bagga and Baldwin, 1998) , canonicalization (Culotta et al., 2007) , and paraphrasing (Barzilay and Lee, 2003) .... | 0 |
In this paper, we explore the application of multilingual learning to part-of-speech tagging when no annotation is available. This core task has been studied in an unsupervised monolingual framework for over a decade and is still an active area of research. In this paper, we demonstrate the effectiveness of multilingua... | 0 |
The rapidly increasing popularity of Social Networking Services (SNSs) such as Twitter and Facebook has greatly eased the dissemination of information. Such data can serve as a valuable information resource for various applications. For instance, Huberman et al. (2009) investigated actual linked structures of human net... | 0 |
We are trying to obtain paraphrases which can be used for Information Extraction (IE) systems. IE systems scan articles and retrieve specific information which is required for a certain domain defined in advance. Currently, many IE tasks are performed by pattern matching. For example, if the system receives a sentence ... | 0 |
Dependency parsers have become a standard module in language technology program pipelines, providing structural information for higher-level tasks such as Information Extraction (Gamallo & Garcia 2012) and Machine Translation (Xu et al. 2009) . Dependency links are computationally easy to process because they are token... | 0 |
The introduction of neural models into natural language processing in the last decade has led to huge improvements in all supervised generation tasks, including morphological inflection. 1 In particular, previous works Silfverberg and Hulden, 2018) have achieved near-perfect performance over the Paradigm Cell Filling P... | 0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.