node_id int64 0 76.9k | label int64 0 39 | text stringlengths 13 124k | neighbors listlengths 0 3.32k | mask stringclasses 4
values |
|---|---|---|---|---|
1,780 | 4 | Using Handhelds as Controls for Everyday Appliances: A Paper Prototype Study Everyday appliances, including telephones, copiers, and home stereos, increasingly contain embedded computers which enable greater functionality. If the interfaces to these appliances were easy to use, people might benefit from these new functions. Unfortunately, it is rare to find a well-designed appliance interface. This study shows that existing appliance interfaces could be improved by using a remote control interface on a handheld computer. Keywords: Handheld computers, remote control, appliance, Pebbles INTRODUCTION The problem with many appliances is that they are too complex. Some appliances need thirty or more buttons to cover all of their functions. This complexity can even make relatively simple tasks, like setting the clock on a VCR, so difficult that people avoid them. Most appliances do not provide unambiguous feedback to users. Indicators of appliance state can be confusing. For example, on a stereo that combines a CD and tape player, it may be difficult to decide wh... | [
2155
] | Validation |
1,781 | 4 | An Overview of the Tatami Project This paper describes the Tatami project at UCSD, which is developing a system to support distributed cooperative software development over the web, and in particular, the validation of concurrent distributed software. The main components of our current prototype are a proof assistant, a generator for documentation websites, a database, an equational proof engine, and a communication protocol to support distributed cooperative work. We believe behavioral specification and verification are important for software development, and for this purpose we use first order hidden logic with equational atoms. The paper also briefly describes some novel user interface design methods that have been developed and applied in the project | [
1761
] | Train |
1,782 | 2 | Using HTML Formatting to Aid in Natural Language Processing on the World Wide Web Because of its magnitude and the fact that it is not computer understandable, the World Wide Web has become a prime candidate for automatic natural language tasks. This thesis argues that there is information in the layout of a web page, and that by looking at the HTML formatting in addition to the text on a page, one can improve performance in tasks such as learning to classify segments of documents. A rich representation for web pages, the HTML Struct Tree, is described. A parsing algorithm for creating Struct Trees is presented, as well as a set of experiments that use Struct Trees as a feature set for learning to extract a company's name and location from its Web pages. Through these experiments we found that it is useful to consider the layout of a page for these tasks. Contents 1. Acknowledgements 2 2. Motivations 3 3. Related Work 3 4. HTML Struct Trees 4 4.1. The Representation : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 4 4.2. The Parsing Algorithm : : :... | [
605,
759,
1676
] | Validation |
1,783 | 1 | Theory of Neuromata A finite automaton --- the so-called neuromaton, realized by a finite discrete recurrent neural network, working in parallel computation mode, is considered. Both the size of neuromata (i.e., the number of neurons) and their descriptional complexity (i.e., the number of bits in the neuromaton representation) are studied. It is proved that a constant time delay of the neuromaton output does not play a role within a polynomial descriptional complexity. It is shown that any regular language given by a regular expression of length n is recognized by a neuromaton with \Theta(n) neurons. Further, it is proved that this network size is, in the worst case, optimal. On the other hand, generally there is not an equivalent polynomial length regular expression for a given neuromaton. Then, two specialized constructions of neural acceptors of the optimal descriptional complexity \Theta(n) for a single n--bit string recognition are described. They both require O(n 1 2 ) neurons and either O(n) con... | [
905,
2976
] | Validation |
1,784 | 2 | Inverted files and dynamic signature files for optimisation of Web Directories Web directories are taxonomies for the classification of Web documents. This kind of IR systems present a specific type of search where the document collection is restricted to one area of the category graph. This paper introduces a specific data architecture for Web directories which improves the performance of restricted searches. That architecture is based on a hybrid data structure composed of an inverted file with multiple embedded signature files. Two variants based on the proposed model are presented: hybrid architecture with total information and hybrid architecture with partial information. The validity of this architecture has been analysed by means of developing both variants to be compared with a basic model. The performance of the restricted queries was clearly improved, specially the hybrid model with partial information, which yielded a positive response under any load of the search system. | [
523,
2503,
2836
] | Train |
1,785 | 0 | Computationally Grounded Theories of Agency In this paper, I motivate, define, and illustrate the notion of computationally grounded theories of agency. A theory of agency is said to be computationally grounded if we can give the theory an interpretation in terms of some concrete computational model. This requirement is essential if we are to claim that the theories we develop can be understood as expressing properties of real multiagent systems. After introducing and formally defining the concept of a computationally grounded theory of agency, I illustrate the idea with reference to VSK logic, a formalism for reasoning about agent systems that has a semantics defined with respect to an automata-like model of agents. VSK logic is an extension of modal epistemic logic, which allows us to represent what information is visible to an agent, what it sees, and what it knows. We are able to prove that formulae of VSK logic correspond directly to properties of agents. 1 Introduction Artificial Intelligence (AI) is a broad church, whic... | [
1899
] | Train |
1,786 | 3 | Probabilistic Object Bases There are many applications where an object oriented data model is a good way of representing and querying data. However, current object database systems are unable to handle the case of objects whose attributes are uncertain. In this paper, extending previous pioneering work by Kornatzky and Shimony, we develop an extension of the relational algebra to the case of object bases with uncertainty. We propose concepts of consistency for such object bases, together with an NP-completeness result, and classes of probabilistic object bases for which consistency is polynomially checkable. In addition, as certain operations involve conjunctions and disjunctions of events, and as the probability of conjunctive and disjunctive events depends both on the probabilities of the primitive events involved as well as on what is known (if anything) about the relationship between the events, we show how all our algebraic operations may be performed under arbitrary probabilistic conjunction and ... | [
2500
] | Train |
1,787 | 3 | Querying Network Directories Hierarchically structured directories have recently proliferated with the growth of the Internet, and are being used to store not only address books and contact information for people, but also personal pro les, network resource information, and network and service policies. These systems provide a means for managing scale and heterogeneity, while allowing for conceptual unity and autonomy across multiple directory servers in the network, in a way far superior to what conventional relational or object-oriented databases o er. Yet, in deployed systems today, much of the data is modeled in an ad hoc manner, and many of the more sophisticated \queries " involve navigational access. In this paper, we develop the core of a formal data model for network directories, and propose a sequence of e ciently computable query languages with increasing expressive power. The directory data model can naturally represent rich forms of heterogeneity exhibited in the real world. Answers to queries expressible in our query languages can exhibit the same kinds of heterogeneity. We present external memory algorithms for the evaluation of queries posed in our directory query languages, and prove the e ciency of each algorithm in terms of its I/O complexity. Our data model and query languages share the exibility and utility of the recent proposals for semi-structured data models, while at the same time e ectively addressing the speci c needs of network directory applications, which we demonstrate by means of a representative real-life example. This work was done when the authors were at AT&T Labs{ | [
655,
891,
1663,
1672,
2876,
3141
] | Test |
1,788 | 1 | Learning Reactive Robot Behaviors with a Neural-Q Learning Approach The purpose of this paper is to propose a Neural-Q_learning approach designed for online learning of simple and reactive robot behaviors. In this approach, the Q_function is generalized by a multi-layer neural network allowing the use of continuous states and actions. The algorithm uses a database of the most recent learning samples to accelerate and guarantee the convergence. Each Neural-Q_learning function represents an independent, reactive and adaptive behavior which maps sensorial states to robot control actions. A group of these behaviors constitutes a reactive control scheme designed to fulfill simple missions. The paper centers on the description of the Neural-Q_learning based behaviors showing their performance with an autonomous underwater vehicle (AUV) in a target following mission. Simulated experiments demonstrate the convergence and stability of the learning system, pointing out its suitability for online robot learning. Advantages and limitations are discussed. | [
470,
961,
1571
] | Test |
1,789 | 5 | Adjustable Autonomy for a Plan Management Agent The Plan Management Agent (PMA) is an intelligent software system that is intended to aid a user in managing a potentially large and complex set of plans. Currently under development, PMA applies AI technology for modeling and reasoning about plans and processes to the development of automated support for work activities. We have developed and implemented algorithms for reasoning about richly expressive plans, which include explicit temporal constraints, temporal uncertainty, and observation actions and conditional branches. We have also developed and implemented an approach to computing the cost of a new plan in the context of existing commitments. The current version of PMA has a low level of autonomy: it makes suggestions to its user, but it does not directly act on her behalf. In this paper, we first describe the PMA system, and then briefly raise some design questions we will need to address as we increase the level of PMA's autonomy, and have it vary with the situation. Introduc... | [
1342,
2364
] | Train |
1,790 | 3 | Tribeca: A System for Managing Large Databases of Network Traffic The engineers who analyze tra c on high bandwidth networks must lter and aggregate either recorded traces of network packets or live tra c from the network itself. These engineers perform operations similar to database queries, but cannot use conventional data managers because of performance concerns and a semantic mismatch between the analysis operations and the operations supported by commercial DBMSs. Tra c analysis does not require fast random access, transactional update, or relational joins. Rather, it needs fast sequential access to a stream of tra c records and the ability to lter, aggregate, de ne windows, demultiplex, and remultiplex the stream. Tribeca is an extensible, stream-oriented DBMS designed to support network tra c analysis. It combines ideas from temporal and sequence databases with an implementation optimized for databases stored on high speed ID-1 tapes or arriving in real time from the network. The paper describes Tribeca's query language, executor and optimizer as well as performance measurements of a prototype implementation. 1 | [
3081
] | Train |
1,791 | 2 | Improving Response Time by Search Pruning in a Content-Based Image Retrieval System, Using Inverted File Techniques This paper describes several methods for improving query evaluation speed in a content-based image retrieval system (CBIRS). Response time is an extremely important factor in determining the usefulness of any interactive system, as has been demonstrated by human factors studies over the past thirty years. In particular, response times of less than one second are often specified as a usability requirement. It is shown that the use of inverted files facilitates the reduction of query evaluation time without significantly reducing the accuracy of the response. The performance of the system is evaluated using precision vs. recall graphs, which are an established evaluation method in information retrieval (IR), and are beginning to be used by CBIR researchers. KEYWORDS: content-based image retrieval, search pruning, inverted file, response time 1 Introduction Response times in the interaction between computer systems and human users are of great importance to user satisfaction. At present... | [
2897
] | Test |
1,792 | 2 | Improving Retrieval on Imperfect Speech Transcriptions This paper presents the results from adding several forms of query expansion to our retrieval system running on transcriptions of broadcast news from the 1997 TREC-7 spoken document retrieval track. 1 Introduction Retrieving documents which originated as speech is complicated by the presence of errors in the transcriptions. If some method of increasing retrieval performance despite these errors could be found, then even low-accuracy recognisers could be used as part of a successful spoken document retrieval (SDR) system. This paper presents results using four query expansion techniques described in [3] on 8 different sets of transcriptions generated for the 1997 TREC-7 SDR evaluation. The baseline retrieval system and the techniques used for query expansion are described in section 2, the transcriptions on which the experiments were performed in section 3 and results and further discussion are offered in section 4. 2 Retrieval Systems 2.1 Baseline System (BL) Our baseline system ... | [
1372,
1895,
2761
] | Validation |
1,793 | 4 | User-Centered Design and Evaluation of a Real-Time Battlefield Visualization Virtual Environment The ever-increasing power of computers and hardware rendering systems has, to date, primarily motivated the creation of visually rich and perceptually realistic virtual environment (VE) applications. Comparatively very little effort has been expended on the user interaction components of VEs. As a result, VE user interfaces are often poorly designed and are rarely evaluated with users. Although usability engineering is a newly emerging facet of VE development, user-centered design and usability evaluation in VEs as a practice still lags far behind what is needed. This paper presents a structured, iterative approach for the user-centered design and evaluation of VE user interaction. This approach consists of the iterative use of expert heuristic evaluation, followed by formative usability evaluation, followed by summative evaluation. We describe our application of this approach to a real-world VE for battlefield visualization, describe the resulting series of design iterations, and present evidence that this approach provides a cost-effective strategy for assessing and iteratively improving user interaction design in VEs. This paper is among the first to report applying an iterative, structured, user-centered design and evaluation approach to VE user interaction design. | [
1141,
1584,
2009,
2185
] | Validation |
1,794 | 2 | Flexible and Scalable Digital Library Search In this report the development of a specialised search engine for a digital library is described. The proposed system architecture consists of three levels: the conceptual, the logical and the physical level. The conceptual level schema enables by its exposure of a domain specific schema semantically rich conceptual search. The logical level provides a description language to achieve a high degree of flexibility for multimedia retrieval. The physical level takes care of scalable and e#cient persistent data storage. The role, played by each level, changes during the various stages of a search engine's lifecycle: (1) modeling the index, (2) populating and maintaining the index and (3) querying the index. The integration of all this functionality allows the combination of both conceptual and content-based querying in the query stage. A search engine for the Australian Open tennis tournament website is used as a running example, which shows the power of the complete architecture and its various components. | [
1014,
2600
] | Test |
1,795 | 0 | On Maintaining Code Mobility We introduce the aspect of maintenance to code mobility with its major problem of keeping track of code migrating through computer networks. Our approach introduces the concept of mobile, lightweight knowledge repositories to support the maintenance of applications deploying mobile code in highly distributed computing environments. Our proposed system establishes a virtual global database with information about the mobile application and its surrounding environment based on distributed structured XML-based knowledge repositories. These distributed databases provide well-defined structural query and retrieval capabilities and ensure the abstraction from programming languages and proprietary hardware platforms. Finally, we point out several future research issues in the field of maintaining mobile code with respect to automatic maintenance of applications and automatic quality checking among others. | [
275,
580,
940
] | Train |
1,796 | 3 | Joins that Generalize: Text Classification Using WHIRL WHIRL is an extension of relational databases that can perform "soft joins" based on the similarity of textual identifiers; these soft joins extend the traditional operation of joining tables based on the equivalence of atomic values. This paper evaluates WHIRL on a number of inductive classification tasks using data from the World Wide Web. We show that although WHIRL is designed for more general similaritybased reasoning tasks, it is competitive with mature inductive classification systems on these classification tasks. In particular, WHIRL generally achieves lower generalization error than C4.5, RIPPER, and several nearest-neighbor methods. WHIRL is also fast---up to 500 times faster than C4.5 on some benchmark problems. We also show that WHIRL can be efficiently used to select from a large pool of unlabeled items those that can be classified correctly with high confidence. Introduction Consider the problem of exploratory analysis of data obtained from the Internet. Assuming that o... | [
630,
1360,
1843,
2780
] | Train |
1,797 | 3 | An Efficient Method for Performing Record Deletions and Updates Using Index Scans We present a method for efficiently performing deletions and updates of records when the records to be deleted or updated are chosen by a range scan on an index. The traditional method involves numerous unnecessary lock calls and traversals of the index from root to leaves, especially when the qualifying records' keys span more than one leaf page of the index. Customers have suffered performance losses from these inefficiencies and have complained about them. Our goal was to minimize the number of interactions with the lock manager, and the number of page fixes, comparison operations and, possibly, I/Os. Some of our improvements come from increased synergy between the query planning and data manager components of a DBMS. Our patented method has been implemented in DB2 V7 to address specific customer requirements. It has also been done to improve performance on the TPC-H benchmark. 1. | [
397
] | Test |
1,798 | 2 | Language Sensitive Text Classification It is a traditional belief that in order to scale-up to more effective retrieval and access methods modern Information Retrieval has to consider more the text content. The modalities and techniques to fit this objectives are still under discussion. More empirical evidence is required to determine the suitable linguistic levels for modeling each IR subtask (e.g. information zoning, parsing, feature selection for indexing,...) and the corresponding use of this information. In this paper an original classification model sensitive to document syntactic information and characterized by a novel inference method is described. Extensive experimental evidence has been derived on real test data and also from well-established academic test sets. The results show that a significant improvement can be derived using the proposed inference model. Also the role of linguistic preprocessing seems to provide positive effects on the performance. POS tagging and recognition of Proper Nouns received a specific experimental attention and provided significant effects on measured accuracy. 1. | [
2838
] | Train |
1,799 | 5 | Coordinating Mobile Robot Group Behavior Using a Model of Interaction Dynamics In this paper we show how various levels of coordinated behavior may be achieved in a group of mobile robots by using a model of the interaction dynamics between a robot and its environment. We present augmented Markov models (AMMs) as a tool for capturing such interaction dynamics on-line and in real-time, with little computational and storage overhead. We begin by describing the structure of AMMs and the algorithm for generating them, then verify the approach utilizing data from physical mobile robots performing elements of a foraging task. Finally, we demonstrate the application of the model for resolving group coordination issues arising from three sources: individual performance, group affiliation, and group performance. Corresponding respectively to these are the three experimental examples we present --- fault detection, group membership based on ability and experience, and dynamic leader selection. 1 Introduction Learning models of the environment, other robots, and interact... | [
1622,
1875,
3080
] | Train |
1,800 | 5 | Hitch hiker's Guide to ESQIMO - Computational model for analogy solving in IQ-tests ESQIMO is a computational model for analogy solving based on a topological formalism of representation. The source and the target analogs are represented as simplexes and the analogy solving is modeled as a topological deformation of these simplexes along a polygonal chain and according to some constraints. We apply this framework to the resolution of IQ-tests typically presented as igiven A, B and C, ønd D such that A is to B what C is to Dj. Keywords : Knowledge Representation, Diagrammatic Reasoning, Simplicial complexes, Analogy solving, CAT. 4 Introduction ESQIMO is a system that automatically solves the following type of IQ-tests: Given three øgures A, B and C, ønd a øgure D such that D is to C what B is to A. These IQ-tests are a paradigmatic example of analogy solving problems. ESQIMO is thus a computational model for analogy solving. The framework used by ESQIMO to solve analogies is called CAT for Combinatorial Algebraic Topology. More precisely, ESQIMO models knowledg... | [
756
] | Train |
1,801 | 3 | Intelligent Techniques for the Extraction and Integration of Heterogeneous Information Developing intelligent tools for the integration of information extracted from multiple heterogeneous sources is a challenging issue to effectively exploit the numerous sources available on-line in global information systems. In this paper, we propose intelligent, tool-supported techniques to information extraction and integration which take into account both structured and semistructured data sources. An object-oriented language called ODL I 3 , derived from the standard ODMG, with an underlying Description Logics, is introduced for information extraction. ODL I 3 descriptions of the information sources are exploited first to set a shared vocabulary for the sources. Information integration is performed in a semiautomatic way, by exploiting ODL I 3 descriptions of source schemas with a combination of Description Logics and clustering techniques. Techniques described in the paper have been implemented in the MOMIS system, based on a conventional mediator architecture. Keywords - Hetero... | [
1238,
1819,
2039,
2235
] | Train |
1,802 | 4 | Model for Unistroke Writing Time Unistrokes are a viable form of text input in pen-based user interfaces. However, they are a very heterogeneous group of gestures the only common feature being that all are drawn with a single stroke. Several unistroke alphabets have been proposed including the original Unistrokes, Graffiti, Allegro, T-Cube and MDITIM. Comparing these methods usually requires a lengthy study with many writers and even then the results are biased by the earlier handwriting experience that the writers have. Therefore, a simple descriptive model can make these comparisons easier. In this paper we propose a model for predicting the writing time for an expert user on any given unistroke alphabet thus enabling sounder argumentation on the properties of different writing methods. Keywords Modeling of motor performance, handwriting, pen input INTRODUCTION Unistrokes were introduced as a text input method for penbased user interfaces by Goldberg and Richardson in their 1993 paper [8]. Unistrokes are an alte... | [
660
] | Train |
1,803 | 3 | Optimizing OID Indexing Cost in Temporal Object-Oriented Database Systems In object-oriented database systems (OODB) with logical OIDs, an OID index (OIDX) is needed to map from OID to the physical location of the object. In a transaction time temporal OODB, the OIDX should also index the object versions. In this case, the index entries, which we call object descriptors (OD), also include the commit timestamp of the transaction that created the object version. In this report, we develop an analytical model for OIDX access costs in temporal OODBs. The model includes the index page buffer as well as an OD cache. We use this model to study access cost and optimal use of memory for index page buffer and OD cache, with different access patterns. The results show that 1) the OIDX access cost can be high, and can easy become a bottleneck in large temporal OODBs, 2) the optimal OD cache size can be relatively large, and 3) the gain from using an optimal size is considerable, and because access pattern in a database system can be very dynamic, the system should be ab... | [
856,
1121,
1187,
1441
] | Train |
1,804 | 2 | Towards a Highly-Scalable Metasearch Engine The World Wide Web has been expanding in a very fast rate. The coverage of the Web by each of the major search engines has been steadily decreasing despite their effort to index more web pages. Worse yet, as these search engines get larger, higher percentages of their indexed information are becoming obsolete. More and more people are having doubt about the scalability of centralized search engine technology. A more scalable alternative to search the Web is the metasearch engine approach. A metasearch engine can be considered as an interface on top of multiple local search engines to provide uniform access to many local search engines. Database selection is one of the main challenges in building a large-scale metasearch engine. The problem is to efficiently and accurately determine a small number of potentially useful local search engines to invoke for each user query. In order to enable accurate selection, metadata that reflect the content of each search engine need to be co... | [
488,
521,
587,
982,
1059,
1120,
1134,
1642,
1888,
2275,
2464,
2503,
2920
] | Test |
1,805 | 0 | Requirements for an Architecture for Embodied Conversational Characters In this paper we describe the computational and architectural requirements for systems which support real-time multimodal interaction with an embodied conversational character. We argue that the three primary design drivers are real-time multithreaded entrainment, processing of both interactional and propositional information, and an approach based on a functional understanding of human face-toface conversation. We then present an architecture which meets these requirements and an initial conversational character that we have developed who is capable of increasingly sophisticated multimodal input and output in a limited application domain. 1 Introduction Research in computational linguistics, multimodal interfaces, computer graphics, and autonomous agents has led to the development of increasingly sophisticated autonomous or semi-autonomous virtual humans over the last five years. Autonomous self-animating characters of this sort are important for use in production animation, interfa... | [
2094,
2537
] | Test |
1,806 | 5 | Using Probabilistic Knowledge and Simulation to Play Poker Until recently, artificial intelligence researchers who use games as their experimental testbed have concentrated on games of perfect information. Many of these games have been amenable to brute-force search techniques. In contrast, games of imperfect information, such as bridge and poker, contain hidden information making similar search techniques impractical. This paper describes recent progress in developing a high-performance pokerplaying program. The advances come in two forms. First, we introduce a new betting strategy that returns a probabilistic betting decision, a probability triple, that gives the likelihood of a fold, call or raise occurring in a given situation. This component unifies all the expert knowledge used in the program, does a better job of representing the type of decision making needed to play strong poker, and improves the way information is propagated throughout the program. Second, real-time simulations are used to compute the expected values of betting decisions. The program generates an instance of the missing data, subject to any constraints that have been learned, and then simulates the rest of the game to determine a numerical result. By repeating this a sufficient number of times, a statistically meaningful sample is used in the program’s decision–making process. Experimental results show that these enhancements each represent major advances in the strength of computer poker programs. 1. | [
713
] | Train |
1,807 | 1 | On the Scalability of Simple Genetic Algorithms Scalable evolutionary computation has become an intensively studied research topic in recent years. The issue of scalability is predominant in any eld of algorithmic design, but it became particularly relevant for the design of competent genetic algorithms once the scalability problems of simple genetic algorithms were understood. Here we present some of the work that has aided in getting a clear insight in the scalability problems of simple genetic algorithms. Particularly, we discuss the important issue of building block mixing and show how the need for mixing places a boundary in the GA parameter space that together with the boundary from the schema theorem delimits the region where the GA converges reliably to the optimum of problems of bounded diculty. This region - or sweet spot as it has been called - shrinks unfortunately very rapidly with increasing problem size unless the building blocks are tightly linked in the problem-coding structure. In addition we look how straightforw... | [
1848
] | Train |
1,808 | 3 | Distributed Knowledge Networks Distributed Knowledge Networks (DKN) provide some of the key enabling technologies for translating recent advances in automated data acquisition, digital storage, computers and communications into fundamental advances in organizational decision support, data analysis, and related applications. DKN include computational tools for accessing, organizing, transforming, and analyzing the contents of heterogeneous, distributed data and knowledge sources and for distributed problem solving and decision making under tight time, resource, and performance constraints. This paper presents an overview of the DKN project in the Iowa State University Artificial Intelligence Laboratory. I. Introduction Advanced scientific research (e.g., the genome project), military applications (e.g., intelligence data handling, situation assessment, command and control) , law enforcement (e.g., terrorism prevention), crisis management, design and manufacturing systems, and medical information infrastructure, pow... | [
551,
1903,
1915,
2046,
2652,
3140
] | Validation |
1,809 | 4 | Bare-Hand Human-Computer Interaction In this paper, we describe techniques for barehanded interaction between human and computer. Barehanded means that no device and no wires are attached to the user, who controls the computer directly with the movements of his/her hand. Our approach is centered on the needs of the user. We therefore define requirements for real-time barehanded interaction, derived from application scenarios and usability considerations. Based on those requirements a finger-finding and hand-posture recognition algorithm is developed and evaluated. To demonstrate the strength of the algorithm, we build three sample applications. Finger tracking and hand posture recognition are used to paint virtually onto the wall, to control a presentation with hand postures, and to move virtual items on the wall during a brainstorming session. We conclude the paper with user tests, which were conducted to prove the usability of bare-hand human computer interaction. Categories and Subject Descriptors H.5.2 [Information interfaces and presentation]: User interfaces -Input devices and strategies; I.5.5 [Pattern recognition]: Implementation - Interactive systems. General Terms Algorithms, Design, Experimentation, Human Factors Keywords Computer Vision, Human-computer Interaction, Real-time, Finger Tracking, Hand-posture Recognition, Bare-hand Control. 1. | [] | Train |
1,810 | 4 | Utility Theory-Based User Models for Intelligent Interface Agents . An underlying problem of current interface agent research is the failure to adequately address effective and efficient knowledge representations and associated methodologies suitable for modeling the users' interactions with the system. These user models lack the representational complexity to manage the uncertainty and dynamics involved in predicting user intent and modeling user behavior. A utility theory-based approach is presented for effective user intent prediction by incorporating the ability to explicitly model users' goals, the uncertainty in the users' intent in pursuing these goals, and the dynamics of users' behavior. We present an interface agent architecture, CIaA, that incorporates our approach and discuss the integration of CIaA with three disparate domains --- a probabilistic expert system shell, a natural language input database query system, and a virtual space plane ---that are being used as test beds for our interface agent research. Keywords: cognitive modeling,... | [
1382,
1413,
2584,
2738
] | Train |
1,811 | 2 | HySpirit - a Probabilistic Inference Engine for Hypermedia Retrieval in Large Databases . HySpirit is a retrieval engine for hypermedia retrieval integrating concepts from information retrieval (IR) and deductive databases. The logical view on IR models retrieval as uncertain inference, for which we use probabilistic reasoning. Since the expressiveness of classical IR models is not sufficient for hypermedia retrieval, HySpirit is based on a probabilistic version of Datalog. In hypermedia retrieval, different nodes may contain contradictory information; thus, we introduce probabilistic four-valued Datalog. In order to support fact queries as well as contentbased retrieval, HySpirit is based on an open world assumption, but allows for predicate-specific closed world assumptions. For performing efficient retrieval on large databases, our system provides access to external data. We demonstrate the application of HySpirit by giving examples for retrieval on images, structured documents and large databases. 1 Introduction Due to the advances in hardware, processing of multimed... | [] | Test |
1,812 | 0 | Dynamic Belief Analysis The process of rational inquiry can be defined as the evolution of the beliefs of a rational agent as a consequence of its internal inference procedures and its interaction with the environment. These beliefs can be modelled in a formal way using belief logics. The possible worlds model and its associated Kripke semantics provide an intuitive semantics for these logics, but they commit us to model agents that are logically omniscient and perfect reasoners. These problems can be avoided with a syntactic view of possible worlds, defining them as arbitrary sets of sentences in a propositional belief logic. In this article this syntactic view of possible worlds is taken, and a dynamic analysis of the beliefs of the agent is suggested in order to model the process of rational inquiry in which the agent is permanently engaged. 1 INTRODUCTION The aim of this work 1 is to model the process of rational inquiry, i.e. the (rationally controlled) transformation of the beliefs of an intelligent... | [
1981,
2240
] | Test |
1,813 | 5 | Issues with Meta-Knowledge This article approaches these questions from our di#erent viewpoints a : The Constructors: Motta and Kalfoglou focus on the construction of knowledge base systems (KBS) using two special kinds of meta-knowledge: ontologies [14,33] and problem solving methods (PSMs) [26,6]. PSMs model the useful inference patterns seen in previous applications. Such patterns, it is argued, simplify and clarify future implementations. Ontologies model common domain terminology. This terminology might include the data structures required by a PSM. Using a good ontology, it is argued, can guide developers in the construction of new systems. For more information on ontologies, see http://www.dai.ed.ac.uk/daidb/people/homes/yannisk/seke99panelhtml.html. The Maintainers: Altho# and Menzies focus on the maintenance and modification of KBS using case-based reasoning (CBR) [3,1] and continual testing [10,22]. CBR researchers argue that people | [
1138
] | Validation |
1,814 | 2 | A Comparative Study of Classification Based Personal E-mail Filtering . This paper addresses personal E-mail filtering by casting it in the framework of text classification. Modeled as semi-structured documents, Email messages consist of a set of fields with predefined semantics and a number of variable length free-text fields. While most work on classification either concentrates on structured data or free text, the work in this paper deals with both of them. To perform classification, a naive Bayesian classifier was designed and implemented, and a decision tree based classifier was implemented. The design considerations and implementation issues are discussed. Using a relatively large amount of real personal E-mail data, a comprehensive comparative study was conducted using the two classifiers. The importance of different features is reported. Results of other issues related to building an effective personal E-mail classifier are presented and discussed. It is shown that both classifiers can perform filtering with reasonable accuracy. While the decision tree based classifier outperforms the Bayesian classifier when features and training size are selected optimally for both, a carefully designed naive Bayesian classifier is more robust. 1 | [
2424
] | Train |
1,815 | 2 | Towards Web-Scale Web Archeology Web-scale Web research is difficult. Information on the Web is vast in quantity, unorganized and uncatalogued, and available only over a network with varying reliability. Thus, Web data is difficult to collect, to store, and to manipulate efficiently. Despite these difficulties, we believe performing Web research at Web-scale is important. We have built a suite of tools that allow us to experiment on collections that are an order of magnitude or more larger than are typically cited in the literature. Two key components of our current tool suite are a fast, extensible Web crawler and a highly tuned, in-memory database of connectivity information. A Web page repository that supports easy access to and storage for billions of documents would allow us to study larger data sets and to study how the Web evolves over time. | [
20,
1170,
1201,
1838,
2372,
2503,
2699
] | Train |
1,816 | 1 | Switch Packet Arbitration via Queue-Learning In packet switches, packets queue at switch inputs and contend for outputs. The contention arbitration policy directly affects switch performance. The best policy depends on the current state of the switch and current traffic patterns. This problem is hard because the state space, possible transitions, and set of actions all grow exponentially with the size of the switch. We present a reinforcement learning formulation of the problem that decomposes the value function into many small independent value functions and enables an efficient action selection. 1 | [
1448
] | Train |
1,817 | 0 | G'day Mate. Let me Introduce you to Everyone: An Infrastructure for Scalable Human-System Interaction We are exposed to physical and virtual systems every day. They consist of computers, PDAs, wireless devices and increasingly, robots. Each provides services to individual or groups of users whether they are local or remote to the system. Services offered by these systems may be useful beyond these users to others, however connecting many of these systems to more users presents a challenging problem. The primary goal of the research presented in this paper is to demonstrate a scalable approach for connecting multiple users to the services provided by multiple systems. Such an approach must be simple, robust and general to contend with the heterogeneous capabilities of the services. An infrastructure is presented that addresses these scalability requirements and establishes the foundation for contending with heterogeneous services. Additionally, it allows services to be linked to form higher-level abstractions. The infrastructure is demonstrated in simulation on several similar multirobot systems with multiple users. The results propose it as a solution for large-scale human-system interaction. | [
1562,
2426
] | Validation |
1,818 | 3 | Active Information Gathering in InfoSleuth . InfoSleuth 1 is an agent-based system that can be configured to perform many different information management activities in a distributed environment. InfoSleuth agents provide a number of complex query services that require resolving ontology-based queries over dynamically changing, distributed, heterogeneous resources. These include distributed query processing, location-independent single-resource updates, event and information monitoring, statistical or inferential data analysis, and trend discovery in complex event streams. It has been used in numerous applications, including the Environmental Data Exchange Network and the Competitive Intelligence System. 1 Introduction In the past 15-20 years, numerous products and prototypes have regularly appeared to provide uniform access to heterogeneous data sources. As a result, that access to heterogeneous sources is now taken as a "given" by customers. Current MCC studies indicate that, given the availability of products that achieve... | [
297,
1552,
3122
] | Validation |
1,819 | 3 | Semantic Integration of Semistructured and Structured Data Sources this paper is to describe the MOMIS [4, 5] (Mediator envirOnment for Multiple Information Sources) approach to the integration and query of multiple, heterogeneous information sources, containing structured and semistructured data. MOMIS has been conceived as a joint collaboration between University of Milano and Modena in the framework of the INTERDATA national research project, aiming at providing methods and tools for data management in Internet-based information systems. Like other integration projects [1, 10, 14], MOMIS follows a "semantic approach" to information integration based on the conceptual schema, or metadata, of the information sources, and on the following architectural elements: i) a common object-oriented data model, defined according to the ODL I 3 language, to describe source schemas for integration purposes. The data model and ODL I 3 have been defined in MOMIS as subset of the ODMG-93 ones, following the proposal for a standard mediator language developed by the I | [
55,
326,
444,
640,
1062,
1481,
1677,
1801,
2039,
2237,
3100
] | Train |
1,820 | 4 | MobiMine: Monitoring the Stock Market from a PDA This paper describes an experimental mobile data mining system that allows intelligent monitoring of time-critical financial data from a hand-held PDA. It presents the overall system architecture and the philosophy behind the design. It explores one particular aspect of the system-automated construction of personalized focus area that calls for user's attention. This module works using data mining techniques. The paper describes the data mining component of the system that employs a novel Fourier analysis-based approach to efficiently represent, visualize, and communicate decision trees over limited bandwidth wireless networks. The paper also discusses a quadratic programming-based personalization module that runs on the PDAs and the multi-media based user-interfaces. It reports experimental results using an ad hoc peer-to-peer IEEE 802.11 wireless network. | [
332,
1401
] | Train |
1,821 | 3 | Community Webs (C-Webs): Technological Assessment and System Architecture this paper, our presentation mainly relies on examples taken from one of the potential C-Web applications, namely C-Web Portals for cultural communities. | [
1146,
1226,
1270,
2100,
2665,
2910,
3101
] | Validation |
1,822 | 3 | Transformation-Based Bottom-Up Computation of the Well-Founded Model We present a bottom-up algorithm for the computation of the well-founded model of non-disjunctive logic programs. Our method is based on the notion of conditional facts and elementary program transformations studied by Bras and Dix [BD94, BD97, BD98b]. However, their "residual program" can grow to exponential size, whereas for function free programs our "program remainder" is always polynomial in the size of the extensional database (EDB). Our approach is also closely related to the alternating fixpoint procedure [VG89, VG93]. However, the alternating fixpoint procedure is known to have inefficiencies due to redundant recomputations of possible facts. By using conditional facts that can be deleted directly instead of recomputing the complement our approach is guaranteed to need not more work than the alternating fixpoint procedure and to be much more efficient in many cases. The magic set transformation as a method for bottom-up query answering is known to have problems with undefined magic facts in the context of the well-founded semantics. However, our approach treats magic set transformed programs correctly, i.e. it always computes a relevant part of the well-founded model of the original program. We show that our approach is guaranteed to need not more work than the well-founded magic sets method [KSS95] or the magic alternating fixpoint procedure [Mor96] and is much more efficient for many programs. Again the use of conditional facts gives valuable information to select the right magic facts to use and to avoid redundant recomputations. | [
1581
] | Train |
1,823 | 2 | Recognizing End-User Transactions in Performance Management Providing good quality of service (e.g., low response times) in distributed computer systems requires measuring end-user perceptions of performance. Unfortunately, in practice such measures are often expensive or impossible to obtain. Herein, we propose a machine learning approach to recognizing end-user transactions consisting of sequences of remote procedure calls (RPCs) received at a server. Two problems are addressed. The first is labeling previously segmented transaction instances with the correct transaction type. This is akin to work done in document classification. The second problem is segmenting RPC sequences into transaction instances. This is a more difficult problem, but it is similar to segmenting sounds into words as in speech understanding. Using Naive Bayes, we tackle the labeling problem with four combinations of feature vectors and probability distributions: RPC occurrences with the Bernoulli distribution and RPC counts with the multinomial, geometric, and shifted ge... | [
759,
1593,
2100
] | Train |
1,824 | 2 | A Content-Based Image Meta-Search Engine using Relevance Feedback Search engines are the most powerful resources for finding information on the rapidly expanding World-Wide Web. Finding the desired search engines and learning how to use them, however, can be very time consuming. Metasearch engines, which integrate a group of such search tools, enable users to access information across the world in a transparent and more efficient manner. The recent emergence of visual information retrieval (VIR) systems on the Web is leading to the same efficiency problem. This paper describes MetaSEEk, a meta-search engine used for retrieving images based on their visual content on the Web. MetaSEEk is designed to intelligently select and interface with multiple on-line image search engines by ranking their performance for different classes of user queries. User feedback is also integrated in the ranking refinement. MetaSEEk has been developed to explore the issues involved in querying large, distributed, on-line visual information system sources. We compare MetaSEE... | [
60,
270,
583,
2275
] | Train |
1,825 | 4 | Distributed and Disappearing User Interfaces in . . . f expression, will often dominate the pure functional aspect to a user. We are already witnessing this transition with mobile phones. Novel computational artifacts based on these technologies may become `invisible', or reside in the background, in several different ways: Truly invisible - when the actual computer and its interface is almost totally integrated with an environment that is familiar to the user. The user interaction model is implicit and perhaps even unnoticeable to the user. Such a system truly resides in the background of the user's attention at all times. Transparent - here the `invisible' UI is not invisible in the literal sense. Rather, it is transparent in the same sense as a very familiar tool is transparent to its user, to the point where it almost acts as an extension of one's body. Every- Distributed and Disappearing User Interfaces in Ubiquitous Computing Keywords context-awareness, disappearing user interfaces | [
692,
1897,
2472
] | Validation |
1,826 | 2 | An Evaluation of Linguistically-motivated Indexing Schemes In this article, we describe a number of indexing experiments based on indexing terms other than simple keywords. These experiments were conducted as one step in validating a linguistically-motivated indexing model. The problem is important but not new. What is new in this approach is the variety of schemes evaluated. It is important since it should not only help to overcome the well-known problems of bag-of-words representations, but also the difficulties raised by non-linguistic text simplification techniques such as stemming, stop-word deletion, and term selection. Our approach in the selection of terms is based on part-of-speech tagging and shallow parsing. The indexing schemes evaluated vary from simple keywords to nouns, verbs, adverbs, adjectives, adjacent word-pairs, and head-modifier pairs. Our findings apply to Information Retrieval and most of related areas. | [
2051,
2427,
3047
] | Test |
1,827 | 2 | Engineering a multi-purpose test collection for Web retrieval experiments Past research into text retrieval methods for the Web has been restricted by the lack of a test collection capable of supporting experiments which are both realistic and reproducible. The 1.69 million document WT10g collection is proposed as a multi-purpose testbed for experiments with these attributes, in distributed IR, hyperlink algorithms and conventional ad hoc retrieval. WT10g was constructed by selecting from a superset of documents in such a way that desirable corpus properties were preserved or optimised. These properties include: a high degree of inter-server connectivity, integrity of server holdings, inclusion of documents related to a very wide spread of likely queries, and a realistic distribution of server holding sizes. We conrm that WT10g contains exploitable link information using a site (homepage) nding experiment. Our results show that, on this task, Okapi BM25 works better on propagated link anchor text than on full text. Keywords: Web retrieval; Link-based ranking; Distributed information retrieval; Test collections 1 | [
471,
1849,
2503
] | Validation |
1,828 | 3 | Semantic Query Optimization through Abduction and Constraint Handling . The use of integrity constraints to perform Semantic Query Optimization (SQO) in deductive databases can be formalized in a way similar to the use of integrity constraints in Abductive Logic Programming (ALP) and the use of Constraint Handling Rules in Constraint Logic Programming (CLP). Based on this observation and on the similar role played by, respectively, extensional, abducible and constraint predicates in SQO, ALP and CLP, we present a unified framework from which (variants of) SQO, ALP and CLP can be obtained as special instances. The framework relies on a proof procedure which combines backward reasoning with logic programming clauses and forward reasoning with integrity constraints. 1 Introduction Semantic Query Optimization (SQO) in deductive databases uses implicit knowledge coded in Integrity Constraints (ICs) to transform queries into new queries that are easier to evaluate and ideally contain only atoms of extensional predicates. SQO sometimes allows for unsatisfiable... | [
872,
1168
] | Validation |
1,829 | 0 | Agent-Based Social Simulation with Coalitions in Social Reasoning There is a growing belief that the agents' cognitive structures play a central role on the enhancement of predicative capacities of decision-making strategies. This paper analyses and simulates the construction of cognitive social structures in the process of decision making with multiple actors. In this process it is argued that the agent's rational choices may be assessed by its motivations, according to different patterns of social interactions. We first construct an abstract model of social dependence between agents, and define a set of social structures that are easily identifiable according to potential interactions. We then carry out a set of experiments at micro-social levels of analysis, where the agents' cognitive structures are explicitly represented. These experiments indicate that different social dependence structures imply distinct structural patterns of negotiation proposals, which appear to have diverse patterns of complexity in the search space. It is subsequently shown that this observation emerges as an issue of ambiguity in the regulation of different decision-making criteria, relative to motivation-oriented and utility-oriented choices. In the scope of this ambiguity, we finally make some conjectures relative to further analytical and empirical analysis around the relation between patterns of complexity of social structures and decision-making. | [
1022,
1158
] | Train |
1,830 | 1 | Identifying Linkage Groups by Nonlinearity/Non-monotonicity Detection This paper presents and discusses direct linkage identification procedures based on nonlinearity/non-monotonicity detection. The algorithm we propose checks arbitrary nonlinearity/non-monotonicity of fitness change by perturbations in a pair of loci to detect their linkage. We first discuss condition of the linkage identification by nonlinearity check (LINC) procedure (Munetomo & Goldberg, 1998) and its allowable nonlinearity. Then we propose another condition of the linkage identification by nonmonotonicity detection (LIMD) and prove its equality to the LINC with allowable nonlinearity (LINC-AN). The procedures can identify linkage groups for problems with at most order-k difficulty by checking O(2 k ) strings and the computational cost for each string is O(l 2 ) where l is the string length. 1 Introduction The definition of linkage in genetics is `the tendency for alleles of different genes to be passed together from one generation to the next' (Winter, Hickey... | [
479,
1191,
1516
] | Train |
1,831 | 1 | Continuous Conceptual Set Covering: Learning Robot Operators From Examples Continuous Conceptual Set Covering (CCSC) is an algorithm that uses engineering knowledge to learn operator effects from training examples. The program produces an operator hypothesis that, even in noisy and nondeterministic domains, can make good quantitative predictions. An empirical evaluation in the traytilting domain shows that CCSC learns faster than an alternative case-based approach. The best results, however, come from integrating CCSC and the case-based approach. Figure 1. Experimental Set Up 1. INTRODUCTION Initially, the robot knows how to physically execute the tilt_operator. It does not, however, know the effects of the operator. When the tray-tilt operator is executed, the robot tips the tray down 30 from the horizontal in the direction of tilt. The new position of the puck is hard to predict because of uncertainty in the initial conditions (the initial position of the puck and the tilting angle are continuous values subject to measurement error). In additio... | [
1178
] | Train |
1,832 | 5 | Augmented Reality: Linking real and virtual worlds - A new paradigm for interacting with computers A revolution in computer interface design is changing the way we think about computers. Rather than typing on a keyboard and watching a television monitor, Augmented Reality lets people use familiar, everyday objects in ordinary ways. The difference is that these objects also provide a link into a computer network. Doctors can examine patients while viewing superimposed medical images; children can program their own LEGO constructions; construction engineers can use ordinary paper engineering drawings to communicate with distant colleagues. Rather than immersing people in an artificiallycreated virtual world, the goal is to augment objects in the physical world by enhancing them with a wealth of digital information and communication capabilities. KEYWORDS: Augmented Reality, Interactive Paper, Design Space Exploration, Participatory Design INTRODUCTION Computers are everywhere: in the past several decades they have transformed our work and our lives. But the conversion from traditi... | [
576,
2147,
2673
] | Test |
1,833 | 3 | Problem solving in ID-logic with aggregates: some experiments The goal of the LP+ project at the K.U.Leuven is to design an expressive logic, suitable for declarative knowledge representation, and to develop intelligent systems based on Logic Programming technology for solving computational problems using the declarative specications. The ID-logic is an integration of typed classical logic and a denition logic. Dierent abductive solvers for this language are being developed. This paper is a report of the integration of high order aggregates into ID-logic and the consequences on the solver SLDNFA. | [
89,
872
] | Train |
1,834 | 4 | The relations between Technologies for Human Learning and Agents . In this position paper we review the historical emergence of Agents | [
1001,
2229
] | Train |
1,835 | 5 | Common Sense and Maximum Entropy This paper concerns the question of how to draw inferences common sensically from uncertain knowledge. Since the early work of Shore and Johnson, [10], Paris and Vencovsk a, [6], and Csiszár, [1], it has been known that the Maximum Entropy Inference Process is the only inference process which obeys certain common sense principles of uncertain reasoning. In this paper we consider the present status of this result and argue that within the rather narrow context in which we work this complete and consistent mode of uncertain reasoning is actually characterised by the observance of just a single common sense principle (or slogan). | [
941
] | Train |
1,836 | 1 | Guest Editors' Introduction: Machine Learning and Natural Language Introduction: Machine Learning and Natural Language CLAIRE CARDIE cardie@cs.cornell.edu Department of Computer Science, Cornell University, Ithaca, NY 14853-7501 RAYMOND J. MOONEY mooney@cs.utexas.edu Department of Computer Sciences, Taylor Hall 2.124, University of Texas, Austin, TX 787121188 The application of machine learning techniques to natural language processing (NLP) has increased dramatically in recent years under the name of "corpus-based," "statistical," or "empirical" methods. However, most of this research has been conducted outside the traditional machine learning research community. This special issue attempts to bridge this divide by assembling an interesting variety of recent research papers on various aspects of natural language learning -- many from authors who do not generally publish in the traditional machine learning literature -- and presenting them to the readers of Machine Learning. In the last five to ten y | [
2676
] | Validation |
1,837 | 3 | Decomposition of Object-Oriented Database Schemas Based on F-logic, we specify an advanced data model with object-oriented and logic-oriented features. For this model we study the decomposition of a class, the counterpart to the well-known decomposition of a relation scheme under functional dependencies. For this decomposition of a class, the transformation pivoting is used. Pivoting transplants some attributes of the class to a newly generated class. This new class is a subclass of the result class of the so-called pivot attribute. The pivot attribute maintains the link between the original class and the new subclass. We identify the conditions for the output of pivoting being equivalent with its input. Additionally, we show under which conditions a schema with functional dependencies can be recursively transformed into an equivalent one without nonkey functional dependencies. 1 Introduction The theory of database schema design aims at formally characterising \good" schemas and at inventing algorithms to measure and to im... | [
174,
2637
] | Train |
1,838 | 2 | Improved Algorithms for Topic Distillation in a Hyperlinked Environment Abstract This paper addresses the problem of topic distillation on the World Wide Web, namely, given a typ-ical user query to find quality documents related to the query topic. Connectivity analysis has been shown to be useful in identifying high quality pages within a topic specific graph of hyperlinked documents. The essence of our approach is to augment a previous connectivity anal-ysis based algorithm with content analysis. We identify three problems with the existing approach and devise al-gorithms to tackle them. The results of a user evaluation are reported that show an improvement of precision at 10 documents by at least 45 % over pure connectivity anal-ysis. 1 | [
2,
116,
124,
255,
333,
477,
507,
538,
553,
608,
825,
865,
949,
1021,
1108,
1201,
1296,
1307,
1512,
1815,
1976,
2038,
2283,
2437,
2459,
2471,
2565,
2595,
2701,
2705,
3029,
3072,
3090,
3131
] | Train |
1,839 | 0 | Deliberate Normative Agents: Principles and Architecture . In this paper norms are assumed to be useful in agent societies. It is claimed that not only following norms, but also the possibility of `intelligent' norm violation can be useful. Principles for agents that are able to behave deliberately on the basis of explicitly represented norms are identified and an architecture is introduced. Using this agent architecture, norms can be communicated, adopted and used as meta-goals on the agent's own processes. As such they have impact on deliberation about goal generation, goal selection, plan generation and plan selection. 1 Introduction Besides autonomy, an important characteristic of agents is that they can react to a changing environment. However, if the protocols that they use to react to (at least some part of) the environment are fixed, they have no ways to respond to impredictable changes. For instance, if an agent notices that another agent is cheating it cannot switch to another protocol to protect itself. (At least this is ... | [
555,
3177
] | Train |
1,840 | 0 | Training Teams with Collaborative Agents . Training teams is an activity that is expensive, time-consuming, hazardous in some cases, and can be limited by availability of equipment and personnel. In team training, the focus is on optimizing interactions, such as efficiency of communication, conflict resolution and prioritization, group situation awareness, resource distribution and load balancing, etc. This paper presents an agent-based approach to designing intelligent team training systems. We envision a computer-based training system in which teams are trained by putting them through scenarios, which allow them to practice their team skills. There are two important roles that intelligent agents can play; these are as virtual team members and as coach. To carry out these functions, these agents must be equipped with an understanding of the task domain, the team structure, the selected decision-making process and their belief about other team members' mental states. 1 Introduction An integral element of large c... | [
109,
808,
1923,
2470,
2815
] | Train |
1,841 | 4 | Combining Positional Information with Visual Media By integrating visual media with positioning information obtained with our wearable computer, we create new opportunities for using visuals both in the field and at the workstation. The position information we store with each visual is direction, pitch, roll, location, focal length, and zoom. This information allows any system to reconstruct the frustum of the visual, and, if height data is available to reconstruct which parts of the earth are visible in the visual. This enables position based lookup and 3D mosaicing of visuals to reconstruct a 3D model. 1 Introduction We are interested in the association of media files with contextual information gathered by our wearable computer. We have previously explored the combination of audio notes, or reminders, with locations [1]. In this paper we turn our attention to visuals. Classic photographs and videos store only visual information. Recent advances in photo and video technology have increased the amount of information which can be st... | [
66
] | Validation |
1,842 | 1 | Learning of Kick in Artificial Soccer Soccer simulation is a suitable domain for research in artificial intelligence. This paper describes a new ball kicking skill, that uses case based learning. In many situations a single kick command is not sufficient to reach the desired ball movement. Hence a skill is needed, that finds a suitable sequence of kicks. The new kicking skill was developed for the AT Humboldt artificial soccer team. | [
2364,
3057,
3173
] | Train |
1,843 | 3 | A Web-based Information System that Reasons with Structured Collections of Text The degree to which information sources are pre-processed by Web-based information systems varies greatly. In search engines like Altavista, little pre-processing is done, while in "knowledge integration" systems, complex site-specific "wrappers" are used integrate different information sources into a common database representation. In this paper we describe an intermediate between these two models. In our system, information sources are converted into a highly structured collection of small fragments of text. Databaselike queries to this structured collection of text fragments are approximated using a novel logic called WHIRL, which combines inference in the style of deductive databases with ranked retrieval methods from information retrieval. WHIRL allows queries that integrate information from multiple Web sites, without requiring the extraction and normalization of object identifiers that can be used as keys; instead, operations that in conventional databases require equality tests... | [
274,
836,
866,
1455,
1532,
1796,
2226,
2371,
2633,
2864,
3053
] | Train |
1,844 | 1 | Case-Based Learning Algorithms Abstract. Storing and using specific instances improves the performance of several supervised learning algorithms. These include algorithms that learn decision trees, classification rules, and distributed networks. However, no investigation has analyzed algorithms that use only specific instances to solve incremental learning tasks. In this paper, we describe a framework and methodology, called instance-based learning, that generates classification predictions using only specific instances. Instance-based learning algorithms do not maintain a set of abstractions derived from specific instances. This approach extends the nearest neighbor algorithm, which has large storage requirements. We describe how storage requirements can be significantly reduced with, at most, minor sacrifices in learning rate and classification accuracy. While the storage-reducing algorithm performs well on several realworld databases, its performance degrades rapidly with the level of attribute noise in training instances. Therefore, we extended it with a significance test to distinguish noisy instances. This extended algorithm's performance degrades gracefully with increasing noise levels and compares favorably with a noise-tolerant decision tree algorithm. | [
25,
623,
631,
884,
1203,
1424,
1490,
2011,
2124,
2171,
2228,
2285,
2298,
2395,
2489,
2548,
2757,
2882,
2890,
3160
] | Test |
1,845 | 0 | Secure Multi-agent Dynamic Programming based on Homomorphic Encryption and its Application to Combinatorial Auctions This paper presents a secure dynamic programming protocol that utilizes homomorphic encryption. By using this method, multiple agents can solve a combinatorial optimization problem among them without leaking their private information. More specifically, in this method, multiple servers cooperatively perform dynamic programming procedures for solving a combinatorial optimization problem by using the private information sent from agents as inputs. Although the severs can compute the optimal solution correctly, the inputs are kept secret even from the servers. | [
1337
] | Test |
1,846 | 1 | An Incremental Learning Algorithm with Automatically Derived Discriminating Features We propose a new technique which incrementally derive discriminating features in the input space. This technique casts both classification problems (class labels as outputs) and regression problems (numerical values as outputs) into a unified regression problem. The virtual labels are formed by clustering in the output space. We use these virtual labels to extract discriminating features in the input space. This procedure is performed recursively. We organize the resulting discriminating subspace in a coarse-to-fine fashion and store the information in a decision tree. Such an incrementally hierarchical discriminating regression (IHDR) decision tree can be realized as a hierarchical probability distribution model. We also introduce a sample size dependent negativelog -likelihood (NLL) metric to deal with large-sample size cases, small-sample size cases, and unbalanced-sample size cases. This is very essential since the number of training samples per class are different at each internal node of the IHDR tree. We report experimental results for two types of data: face image data along with comparison with some major appearance-based method and decision trees, hall way images with driving directions as outputs for the automatic navigation problem -- a regression application. | [
3058
] | Validation |
1,847 | 2 | Inducing Conceptual User Models Usually, performance is the primary objective in systems that make use of user modeling (Um) techniques. But since machine learning (Ml) in user modeling addresses several issues in the context of human computer interaction (Hci), the requirements on "performance" are manifold. Thus, machine learning for user modeling (Ml4Um) has to meet several demands in order to satisfy the aims of involved disciplines. In this article we describe the application of a rather unusual Ml method to Um, namely inductive logic programming (Ilp). Though not primarily associated with efficient learning methods, we motivate the use of Ilp by showing translucency of derived user models and the explanatory potential of such models during a user adapted filtering process. In course of the OySTER project our goal is to induce conceptual user models that allow for a transparent query refinement and information filtering in the domain of Www meta--search. 1 Introduction: Why Conceptual User Models?... | [
759,
1053,
1510,
3035
] | Test |
1,848 | 1 | BOA: The Bayesian Optimization Algorithm In this paper, an algorithm based on the concepts of genetic algorithms that uses an estimation of a probability distribution of promising solutions in order to generate new candidate solutions is proposed. To estimate the distribution, techniques for modeling multivariate data by Bayesian networks are used. The proposed algorithm identifies, reproduces and mixes building blocks up to a specified order. It is independent of the ordering of the variables in the strings representing the solutions. Moreover, prior information about the problem can be incorporated into the algorithm. However, prior information is not essential. Preliminary experiments show that the BOA outperforms the simple genetic algorithm even on decomposable functions with tight building blocks as a problem size grows. 1 INTRODUCTION Recently, there has been a growing interest in optimization methods that explicitly model the good solutions found so far and use the constructed model to guide the fu... | [
1191,
1384,
1516,
1807
] | Train |
1,849 | 2 | Measuring Search Engine Quality The effectiveness of twenty public search engines is evaluated using TREC-inspired methods and a set of 54 queries taken from real Web search logs. The World Wide Web is taken as the test collection and a combination of crawler and text retrieval system is evaluated. The engines are compared on a range of measures derivable from binary relevance judgments of the first seven live results returned. Statistical testing reveals a significant difference between engines and high inter-correlations between measures. Surprisingly, given the dynamic nature of the Web and the time elapsed, there is also a high correlation between results of this study and a previous study by Gordon and Pathak. For nearly all engines, there is a gradual decline in precision at increasing cutoff after some initial fluctuation. Performance of the engines as a group is found to be inferior to the group of participants in the TREC-8 Large Web task, although the best engines approach the median of those systems. Shortcomings of current Web search evaluation methodology are identified and recommendations are made for future improvements. In particular, the present study and its predecessors deal with queries which are assumed to derive from a need to find a selection of documents relevant to a topic. By contrast, real Web search reflects a range of other information need types which require different judging and different measures. The authors wish to acknowledge that this work was carried out partly within the Cooperative Research Centre for Advanced Computational Systems established under the Australian Government's Cooperative Research Centres Program. 1 1 | [
885,
1827
] | Test |
1,850 | 3 | Data Partitioning and Load Balancing in Parallel Disk Systems Parallel disk systems provide opportunities for exploiting I/O parallelism in two possible ways, namely via inter-request and intra-request parallelism. In this paper we discuss the main issues in performance tuning of such systems, namely striping and load balancing, and show their relationship to response time and throughput. We outline the main components of an intelligent, self-reliant file system that aims to optimize striping by taking into account the requirements of the applications, and performs load balancing by judicious file allocation and dynamic redistributions of the data when access patterns change. Our system uses simple but effective heuristics that incur only little overhead. We present performance experiments based on synthetic workloads and real-life traces. Keywords: parallel disk systems, performance tuning, file striping, data allocation, load balancing, disk cooling. 1 Introduction: Tuning Issues in Parallel Disk Systems Parallel disk systems are of great imp... | [
1416
] | Train |
1,851 | 4 | Stable Haptic Interaction with Virtual Environments A haptic interface is a kinesthetic link between a human operator and a virtual environment. This paper addresses fundamental stability and performance issues associated with haptic interaction. It generalizes and extends the concept of a virtual coupling network, an artificial link between the haptic display and a virtual world, to include both the impedance and admittance models of haptic interaction. A benchmark example exposes an important duality between these two cases. Linear circuit theory is used to develop necessary and sufficient conditions for the stability of a haptic simulation, assuming the human operator and virtual environment are passive. These equations lead to an explicit design procedure for virtual coupling networks which give maximum performance while guaranteeing stability. By decoupling the haptic display control problem from the design of virtual environments, the use of a virtual coupling network frees the developer of haptic-enabled virtual reality models fr... | [
480
] | Validation |
1,852 | 3 | OBSERVER: An Approach for Query Processing in Global Information Systems based on Interoperation across Pre-existing Ontologies . There has been an explosion in the types, availability and volume of data accessible in an information system, thanks to the World Wide Web (the Web) and related inter-networking technologies. In this environment, there is a critical need to replace or complement earlier database integration approaches and current browsing and keyword-based techniques with concept-based approaches. Ontologies are increasingly becoming accepted as an important part of any concept or semantics based solution, and there is increasing realization that any viable solution will need to support multiple ontologies that may be independently developed and managed. In particular, we consider the use of concepts from pre-existing real world domain ontologies for describing the content of the underlying data repositories. The most challenging issue in this approach is that of vocabulary sharing, which involves dealing with the use of different terms or concepts to describe similar information. In this paper, we ... | [
1677,
3092
] | Train |
1,853 | 4 | VRML with Constraints In this paper we discuss the benefits of extending VRML by constraints and present a new way based on prototypes and scripting to implement this extension. Our approach is easy-to-use, extensible and it considerably increases the expressivity of VRML. Our implementation supports one-way equational and finite domain constraints. We demonstrate the use of these constraints by means of several examples. Finally we argue that in the long run constraints should become an integral part of VRML. CR Categories: I.3.7 [Computer Graphics]: Three-Dimensional Graphics and Realism---Virtual Reality; D.3.3 [Programming Languages ]: Language Constructs and Features---Constraints Keywords: VRML, Animation, Programming 1 | [
1918
] | Train |
1,854 | 4 | The Gesture Pendant: A Self-illuminating, Wearable, Infrared Computer Vision System for Home Automation Control and Medical Monitoring In this paper we present a wearable device for control of home automation systems via hand gestures. This solution has many advantages over traditional home automation interfaces in that it can be used by those with loss of vision, motor skills, and mobility. By combining other sources of context with the pendant we can reduce the number and complexity of gestures while maintaining functionality. As users input gestures, the system can also analyze their movements for pathological tremors. This information can then be used for medical diagnosis, therapy, and emergency services.Currently, the Gesture Pendant can recognize control gestures with an accuracy of 95% and user defined gestures with an accuracy of 97% It can detect tremors above 2HZ within /- Hz. | [
187,
261,
727,
1772,
2203
] | Test |
1,855 | 5 | Trends In Evolutionary Robotics A review is given on the use of evolutionary techniques for the automatic design of adaptive robots. The focus is on methods which use neural networks and have been tested on actual physical robots. The chapter also examines the role of simulation and the use of domain knowledge in the evolutionary process. It concludes with some predictions about future directions in robotics. Appeared in Soft Computing for Intelligent Robotic Systems, edited by L.C. Jain and T. Fukuda, PhysicaVerlag, New York, NY, 215--233, 1998. 1 Introduction To be truly useful, robots must be adaptive. They should have a collection of basic abilities that can be brought to bear in tackling a variety of tasks in a wide range of environments. These fundamental abilities might include navigation to a goal location, obstacle avoidance, object recognition, and object manipulation. However, to date, this desired level of adaptability has not been realized. Instead, robots have primarily been successful when deploye... | [
2634
] | Train |
1,856 | 0 | Mobile Objects in Java Mobile Objects in Java provides support for object mobility in Java. Similarly to the RMI technique, a notion of client-side stub, called startpoint, is used to communicate transparently with a server-side stub, called endpoint. Objects and associated endpoints are allowed to migrate. Our approach takes care of routing method calls using an algorithm that we studied in [22]. The purpose of this paper is to present and evaluate the implementation of this algorithm in Java. In particular, two different strategies for routing method invocations are investigated, namely call forwarding and referrals. The result of our experimentation shows that the latter can be more efficient by up to 19%. 1 | [
1682,
3172
] | Train |
1,857 | 2 | WebMate: A Personal Agent for Browsing and Searching The World-Wide Web is developing very fast. Currently, #nding useful information on the Web is a time consuming process. In this paper, we presentWebMate, an agent that helps users to e#ectively browse and search the Web. WebMate extends the state of the art in Web-based information retrieval in manyways. First, it uses multiple TF-IDF vectors to keep track of user interests in di#erent domains. These domains are automatically learned byWebMate. Second, WebMate uses the Trigger Pair Model to automatically extract keywords for re#ning document search. Third, during search, the user can provide multiple pages as similarity#relevance guidance for the search. The system extracts and combines relevantkeywords from these relevant pages and uses them for keyword re#nement. Using these techniques, WebMate provides e#ective browsing and searching help and also compiles and sends to users personal newspaper by automatically spiding news sources. Wehave experimentally evaluated the performance of the system. | [
495,
760,
1377,
1555,
2336,
2589,
2738,
2816
] | Train |
1,858 | 2 | Image Retrieval: Past, Present, And Future This paper provides a comprehensive survey of the technical achievements in the research area of Image Retrieval, especially Content-Based Image Retrieval, an area so active and prosperous in the past few years. The survey includes 100+ papers covering the research aspects of image feature representation and extraction, multi-dimensional indexing, and system design, three of the fundamental bases of Content-Based Image Retrieval. Furthermore, based on the state-of-the-art technology available now and the demand from real-world applications, open research issues are identified, and future promising research directions are suggested. 1. INTRODUCTION Recent years have seen a rapid increase of the size of digital image collections. Everyday, both military and civilian equipment generates giga-bytes of images. Huge amount of information is out there. However, we can not access to or make use of the information unless it is organized so as to allow efficient browsing, searching and retriev... | [
270,
863,
3031
] | Test |
1,859 | 4 | On the Expressiveness of Event Notification in Data-driven Coordination Languages JavaSpaces and TSpaces are two coordination middlewares for distributed Java programming recently proposed by Sun and IBM, respectively. They extend the data-driven coordination model of Linda with an event notification mechanism: a process can register interest in the incoming arrivals of a particular kind of data and receive communication of the occurrences of these events. In [2, 3] we introduce a process calculus L, based on the Linda coordination model, and we prove that if processes are synchronous with the dataspace (ordered interpretation) L is Turing powerful, while this is not the case if they are asynchronous (unordered interpretation). Here, we introduce a new calculus Ln , obtained by extending L with the event noti cation mechanism, and we prove two main results contrasting with what has been shown for L: (i) Ln is Turing powerful also under the unordered interpretation and (ii) it allows a faithful encoding of the ordered semantics on top of the unordered one. | [
3045
] | Test |
1,860 | 4 | The Vesta Approach to Software Configuration Management Vesta is a system for software configuration management. It stores collections of source files, keeps track of which versions of which files go together, and automates the process of building a complete software artifact from its component pieces. Vesta's novel approach gives it three important properties. First, every build is repeatable, because its component sources and build tools are stored immutably and immortally, and its configuration description completely specifies what components and tools are used and how they are put together. Second, every build is incremental, because results of previous builds are cached and reused. Third, every build is consistent, because all build dependencies are automatically captured and recorded, so that a cached result from a previous build is reused only when doing so is certain to be correct. In addition, Vesta's flexible language for writing configuration descriptions makes it easy to describe large software configurations in a modular fashion and to create variant configurations by customizing build parameters. This paper gives a brief overview of Vesta, outlining Vesta's advantages over traditional tools, how those benefits are achieved, and the system's overall performance. | [
1218
] | Validation |
1,861 | 4 | Location and Recovery of Text on Oriented Surfaces We present a method for extracting text from images where the text plane is not necessarily fronto-parallel to the camera. Initially, we locate local image features such as borders and page edges. We then use perceptual grouping on these features to find rectangular regions in the scene. These regions are hypothesised to be pages or planes that may contain text. Edge distributions are then used for the assessment of these potential regions, providing a measure of confidence. It will be shown that the text may then be transformed to a fronto-parallel view suitable, for example, for an OCR system or other higher level recognition. The proposed method is scale independent (of the size of the text). We illustrate the algorithm using various examples. Keywords: Oriented Text, Perspective Recovery of Text, Edge Angle Distribution 1. INTRODUCTION Location and recovery of text in a scene would be useful in the context of wearable computing, desk computing, or unguided robotic motion. Such a... | [
2323,
2938
] | Validation |
1,862 | 0 | A Business Process Agent The architecture of a process agent is a three-layer BDI, hybrid, multi-agent architecture. These process agents are intended to deal with corporate cultural, or political, sensitivities as well as with corporate rules. These agents adapt their behaviour on the basis of the likelihood of plan success, and on estimates of the time, cost and value of choosing a plan. 1 Introduction An intelligent multi-agent system is a society of autonomous cooperating components each of which maintains an ongoing interaction with its environment. Intelligent agents should be autonomous, cooperative and adaptive. A process agent architecture is designed specifically for business process applications. Typically the cost of bringing a business process to its conclusion is substantially due to the cost of the human processing involved. If this is so then a process management system should make its decisions in a thorough and considered way, and should have no reason, based on cost, for not doing s... | [
2364
] | Validation |
1,863 | 1 | Probabilistic Affine Invariants for Recognition Under a weak perspective camera model, the image plane coordinates in different views of a planar object are related by an affine transformation. Because of this property, researchers have attempted to use affine invariants for recognition. However, there are two problems with this approach: (1) objects or object classes with inherent variability cannot be adequately treated using invariants; and (2) in practice the calculated affine invariants can be quite sensitive to errors in the image plane measurements. In this paper we use probability distributions to address both of these difficulties. Under the assumption that the feature positions of a planar object can be modeled using a jointly Gaussian density, we have derived the joint density over the corresponding set of affine coordinates. Even when the assumptions of a planar object and a weak perspective camera model do not strictly hold, the results are useful because deviations from the ideal can be treated as deformability in the ... | [
405,
2430,
3134
] | Test |
1,864 | 0 | Parameter Control Using the Agent Based Patchwork Model The setting of parameters in Evolutionary Algorithms (EA) has crucial influence on their performance. Typically, the best choice depends on the optimization task. Some parameters yield better results when they are varied during the run. Recently, the so-called TerrainBased Genetic Algorithm (TBGA) was introduced, which is a self-tuning version of the traditional Cellular Genetic Algorithm (CGA). In a TBGA, the individuals of the population are placed in a two-dimensional grid, where only neighbored individuals can mate with each other. The position of an individual in this grid is interpreted as its offspring 's specific mutation rate and number of crossover points. This approach allows to apply GA parameters that are optimal for (i) the type of optimization task and (ii) the current state of the optimization process. However, only a few individuals can apply the optimal parameters simultaneously due to their fixed position in the grid lattice. In this paper, we substituted the fixed s... | [
1029
] | Validation |
1,865 | 0 | Coordinating Mobile Agents by means of Communicators This paper proposes a coordination model, for both static and mobile agents, based on abstract structures called Communicators, entities which handle agent dialogue performed through ACL speech act exchanging. Such structures are designed based on a need to model agent dialogue in a human-like style, offering a set of coordination primitives of general validity, able to provide both a direct and indirect interaction model. Since a Communicator handles messages exchanged within a well-defined multi-agent application, it is fully programmable, i.e. it is possible to specify what messages can be exchanged and how these message have to be handled. In detail, a Communicator performs a syntactic and semantic routing, allowing the exchange and forwarding of a message according to the programmed coordination laws. | [
2629
] | Validation |
1,866 | 2 | Annotate - A Web-based Knowledge Management Support System for Document Collections Difficulties with web-based full text information retrieval (IR) systems include spurious matches, manually intensive document sifting, and the absence of communication or coordination between users. Furthermore, it is not clear how to best design a Computer-Supported Collaborative Work (CSCW) system which possesses coordination mechanisms in the inherently stateless Web environment. Document collections in an Intranet are a particularly interesting problem; electronic documents contain the potential for improving information throughput and knowledge transfer in organizations. The emergence of Intranet infrastructure creates new IR demands as all users can easily publish documents online. The associated problems include inefficient discovery and proliferation of documents regardless of quality, which in turn implies the mismanagement of the authors' expertise. This thesis investigates how to design an IR system to support KM in organizations. I propose an architecture and implement a n... | [
499,
3009
] | Train |
1,867 | 0 | Architectural Design Patterns for Multiagent Coordination This paper presents our first step towards agent-oriented software engineering, focusing on the area of coordinated multi-agent systems. In multi-agent systems, the interactions between the agents are crucial in determining the effectiveness of the system. Hence the adoption of an appropriate coordination mechanism is pivotal in the design of multi-agent system architectures. This paper does not focus on agent theory, rather on the development of an agent-oriented software engineering methodology, collaboration architectures and design patterns for collaboration. A catalog of coordination patterns inherent in multi-agent architectures is presented. Such patterns may be utilized in the architectural design for multiagent systems, allowing researchers and practitioners to improve the integrability and reusability properties of their systems. | [
438
] | Train |
1,868 | 1 | Attribute Grammars for Genetic Representations of Neural Networks and Syntactic Constraints of Genetic Programming context-free grammar augmented by the assignment of semantic attributes to the symbols of the grammar. A production rule specifies not only the replacement of symbols, but also the evaluation of the symbol’s attributes. In our research, an attribute grammar is used to specify classes of neural network structures with explicit representation of their functional organization. These representations provide useful constraints upon a genetic optimization that guarantee the preservation of syntactically correct genetic trees with semantically meaningful sub-trees. In this paper, we give a broad overview of our research into attribute grammar representations, from the basic and known capabilities, to the current ideas being addressed, to the future directions of our research. Attribute Grammars and Neural Networks | [
1509
] | Train |
1,869 | 4 | Seeking a Foundation for Context-Aware Computing Context-aware computing is generally associated with elements of the Ubiquitous Computing program, and the opportunity to distribute computation and interaction through the environment rather than concentrating it at the desktop computer. However, issues of context have also been important in other areas of HCI research. I argue that the scope of context-based computing should be extended to include not only Ubiquitous Computing, but also recent trends in tangible interfaces as well as work on sociological investigations of the organization of interactive behavior. By taking a view of contextaware computing that integrates these different perspectives, we can begin to understand the foundational relationships that tie them all together, and that provide a framework for understanding the basic principles behind these various forms of embodied interaction. In particular, I point to phenomenology as a basis for the development of a new framework for design and evaluation of context-aware ... | [
2526
] | Train |
1,870 | 4 | Co-operative Evaluation of a Desktop Virtual Reality System A summative usability evaluation of a desktop virtual reality (VR) system was developed and a preliminary study then conducted. The purpose of the study was twofold. Firstly, to test whether the traditional evaluation technique, co-operative evaluation, is effective in the evaluation of desktop VR systems. Co-operative evaluation is a variation on a `think-aloud' verbal protocol, whereby, in addition to concurrently 'thinking-aloud', users are encouraged to ask any questions about an evaluation, relating to the computer-based system, the application, or the tasks that they are required to perform during the evaluation. As well as this, the evaluator may ask questions of the user at any time during the evaluation. Results from the study indicate that this additional probing technique enables an evaluator to elicit further usability problems that may not have otherwise been exteriorized by the user. Additionally, a method is developed which attempts to turn round the qualitative 'think-aloud' type data into quantitative data. This provides a way of evaluating empirical 'think-aloud' evaluation methods and will be useful for comparing their effectiveness to evaluate 3D virtual reality systems. | [
1276,
1585,
2649
] | Train |
1,871 | 1 | Stochastic Attribute Selection Committees Classifier committee learning methods generate multiple classifiers to form a committee by repeated application of a single base learning algorithm. The committee members vote to decide the final classification. Two such methods, Bagging and Boosting, have shown great success with decision tree learning. They create different classifiers by modifying the distribution of the training set. This paper studies a different approach: Stochastic Attribute Selection Committee learning of decision trees. It generates classifier committees by stochastically modifying the set of attributes but keeping the distribution of the training set unchanged. An empirical evaluation of a variant of this method, namely Sasc, in a representative collection of natural domains shows that the SASC method can significantly reduce the error rate of decision tree learning. On average Sasc is more accurate than Bagging and less accurate than Boosting, although a one-tailed sign-test fails to show that these differences are significant at a level of 0.05. In addition, it is found that, like Bagging, Sasc is more stable than Boosting in terms of less frequently obtaining significantly higher error rates than C4.5 and, when error is raised, producing lower error rate increases. Moreover, like Bagging, Sasc is amenable to parallel and distributed processing while Boosting is not. | [
191,
1896,
2210
] | Validation |
1,872 | 4 | A Virtual Reality based System Environment for Intuitive Walk-Throughs and Exploration of Large-Scale Tourist Information This paper describes the concept and prototype architecture for Virtual Reality (VR) based Information Systems (ViRXIS). ViRXIS may serve as a base architecture for different kind of IS domains, such as a VR based Tourist Information System (ViRTIS) or a VR based Geographic Information System (ViRGIS). Finally, potential application scenarios of ViRTIS will be presented. Keywords Tourist Information Systems, Virtual Reality, virtual worlds, real-time interactive 3D simulation, information systems, man-machine-interface, object-oriented database management systems, spatial data access structures. 1. Introduction The ever increasing computing power and storage capacity of low-cost computer systems enables the implementation of multimedia applications that integrate different media such as text, image, graphics, voice, music, computer animation, or video for the presentation and manipulation of tourist information. At present, the userinterface of such multimedia-based Tourist Informati... | [
2869
] | Train |
1,873 | 1 | Evolution and Development of a Central Pattern Generator for the Swimming of a Lamprey This paper describes the design of neural control architectures for locomotion using an evolutionary approach. Inspired by the central pattern generators found in animals, we develop neural controllers which can produce the patterns of oscillations necessary for the swimming of a simulated lamprey. This work is inspired by Ekeberg's neuronal and mechanical model of a lamprey [11], and follows experiments in which swimming controllers were evolved using a simple encoding scheme [26, 25]. Here, controllers are developed using an evolutionary algorithm based on the SGOCE encoding [31, 32] in which a genetic programming approach is used to evolve developmental programs which encode the growing of a dynamical neural network. The developmental programs determine how neurons located on a 2D substrate produce new cells through cellular division and how they form efferent or afferent interconnections. Swimming controllers are generated when the growing networks eventually create connections to ... | [
926,
2557,
2879
] | Test |
1,874 | 2 | Detecting Concept Drift with Support Vector Machines For many learning tasks where data is collected over an extended period of time, its underlying distribution is likely to change. A typical example is information filtering, i.e. the adaptive classification of documents with respect to a particular user interest. Both the interest of the user and the document content change over time. A filtering system should be able to adapt to such concept changes. This paper proposes a new method to recognize and handle concept changes with support vector machines. The method maintains a window on the training data. The key idea is to automatically adjust the window size so that the estimated generalization error is minimized. The new approach is both theoretically well-founded as well as effective and efficient in practice. Since it does not require complicated parameterization, it is simpler to use and more robust than comparable heuristics. Experiments with simulated concept drift scenarios based on real-world text data com... | [
888
] | Validation |
1,875 | 0 | A Multiagent Framework for Planning, Reacting, and Learning . There are two main approaches to activity coordination in multiagent systems, plan-based coordination and reactive coordination, each having its specic advantages and disadvantages. This paper describes a framework called M-Dyna-Q that aims at combining these two approaches such that their advantages are retained while their disadvantages are avoided. The key idea behind M-Dyna-Q, which is a multiagent variant of a single-agent framework known as Dyna-Q, is to achieve this combination through equipping the agents with the capability to learn information that is relevant to their planning and reacting activities. M-Dyna-Q thus oers an integrated view of planning, reacting, and learning in multiagent systems. Keywords. (M-)Dyna-Q, multiagent systems, reactive coordination, plan-based coordination, learning. 1 Introduction The past years have witnessed a rapidly growing interest in multiagent systems (MAS), that is, in systems in which several interacting, intelligent and autono... | [
1799,
2321
] | Test |
1,876 | 4 | User Modeling in Human-Computer Interaction A fundamental objective of human-computer interaction research is to make systems more usable, more useful, and to provide users with experiences fitting their specific background knowledge and objectives. The challenge in an information-rich world is not only to make information available to people at any time, at any place, and in any form, but specifically to say the right thing at the right time in the right way. Designers of collaborative humancomputer systems face the formidable task of writing software for millions of users (at design time) while making it work as if it were designed for each individual user (only known at use time). User modeling research has attempted to address these issues. In this article, I will first review the objectives, progress, and unfulfilled hopes that have occurred over the last ten years, and illustrate them with some interesting computational environments and their underlying conceptual frameworks. A special emphasis is given to high-functionali... | [
414,
1692,
2895
] | Test |
1,877 | 3 | Progressive KNN Search Using B+-trees In high-dimensional databases, nearest neighbor (NN) search is computationally expensive. However, for many applications where small errors can be tolerated, determining approximate answers quickly has become an acceptable alternative. In this paper, we present a new algorithm that exploits the iMinMax( ) scheme [16] to facilitate nearest neighbor processing. The proposed algorithm distinguishes itself from existing algorithms in two ways. First, it evaluates KNN queries progressively, i.e., it begins by examining a small amount of data to produce approximate answers quickly; as more data are examined, the approximate answers are refined. Second, unlike existing progressive algorithms that access data randomly, the algorithm searches the data space in a systematic and `optimal' manner. Our performance study shows that the proposed scheme can produce approximate answers very quickly. More importantly, the quality of the answers are not sacrified significantly. Such an extension is applicable to other relevant methods such as the Pyramid Technique[2]. | [
547
] | Train |
1,878 | 0 | CABOB: A Fast Optimal Algorithm for Combinatorial Auctions Combinatorial auctions where bidders can bid on bundles of items can lead to more economical allocations, but determining the winners is NP-complete and inapproximable. We present CABOB, a sophisticated search algorithm for the problem. It uses decomposition techniques, upper and lower bounding (also across components) , elaborate and dynamically chosen bid ordering heuristics, and a host of structural observations. Experiments against CPLEX 7.0 show that CABOB is usually faster, never drastically slower, and in many cases with special structure, drastically faster. We also uncover interesting aspects of the problem itself. First, the problems with short bids that were hard for the first-generation of specialized algorithms are easy. Second, almost all of the CATS distributions are easy, and become easier with more bids. Third, we test a number of random restart strategies, and show that they do not help on this problem because the run-time distribution does not have a heavy tail (at least not for CABOB). 1 | [
1337,
1697,
2707
] | Test |
1,879 | 3 | A New Heuristic for Optimizing Large Queries There is a number of OODB optimization techniques proposed recently, such as the translation of path expressions into joins and query unnesting, that may generate a large number of implicit joins even for simple queries. Unfortunately, most current commercial query optimizers are still based on the dynamic programming approach of System R, and cannot handle queries of more than ten tables. There is a number of recent proposals that advocate the use of combinatorial optimization techniques, such as iterative improvement and simulated annealing, to deal with the complexity of this problem. These techniques, though, fail to take advantage of the rich semantic information inherent in the query specification, such as the information available in query graphs, which gives a good handle to choose which relations to join each time. This paper presents a polynomial-time algorithm that generates a good quality order of relational joins. It can also be used with minor modifications to sort OODB a... | [
19,
2575
] | Train |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.