node_id int64 0 76.9k | label int64 0 39 | text stringlengths 13 124k | neighbors listlengths 0 3.32k | mask stringclasses 4
values |
|---|---|---|---|---|
680 | 5 | The Erlangen Spoken Dialogue System EVAR: A State-of-the-Art Information Retrieval System In this paper, we present an overview of the spoken dialogue system EVAR that was developed at the University of Erlangen. In January 1994, it became accessible over telephone line and could answer inquiries in the German language about German InterCity train connections. It has since been continuously improved and extended, including some unique features, such as the processing of out--of--vocabulary words and a flexible dialogue strategy that adapts to the quality of the recognition of the user input. In fact, several different versions of the system have emerged, i.e. a subway information system, train and flight information systems in different languages, and an integrated multilingual and multifunctional system which covers German and 3 additional languages in parallel. Current research focuses on the introduction of stochastic models into the semantic analysis, on the direct integration of prosodic information into the word recognition process, on the detection of user emotion, a... | [
2305
] | Train |
681 | 4 | Adding Some Smartness to Devices and Everyday Things In mobile computing, context-awareness indicates the ability of a system to obtain and use information on aspects of the system environment. To implement contextawareness, mobile system components have to be augmented with the ability to capture aspects of their environment. Recent work has mostly considered locationawareness, and hence augmentation of mobile artifacts with locality. In this paper we discuss augmentation of mobile artifacts with diverse sets of sensors and perception techniques for awareness of context beyond location. We report experience from two projects, one on augmentation of mobile phones with awareness technologies, and the other on embedding of awareness technology in everyday non-digital artifacts. | [
664,
1156,
3049,
3087
] | Train |
682 | 4 | Software Infrastructure for Ubiquitous Computing Environments: Supporting Synchronous Collaboration with Heterogeneous Devices In ubiquitous computing environments, multiple users work with a wide range of different devices. In many cases, users interact and collaborate using multiple heterogeneous devices at the same time. The configuration of the devices should be able to change frequently due to a highly dynamic, flexible and mobile nature of new work practices. This produces new requirements for the architecture of an appropriate software infrastructure. In this paper, an architecture designed to meet these requirements is proposed. To test its applicability, this architecture was used as the basis for the implementation of BEACH, the software infrastructure of i-LAND (the ubiquitous computing environment at GMD-IPSI). It provides the functionality for synchronous cooperation and interaction with roomware components, i.e. room elements with integrated information technology. In conclusion, our experiences with the current implementation are presented. | [
1304,
1433,
1645,
2351,
3088,
3148
] | Train |
683 | 0 | Dynamic User Model Construction with Bayesian Networks for Intelligent Information Queries The complexity of current software applications is overwhelming users. The need exists for intelligent interface agents to address the problems of increasing taskload that is overwhelming the human user. Interface agents could help alleviate user taskload by extracting and analyzing relevant information, and providing information abstractions of that information, and providing timely, beneficial assistance to users. Central to providing assistance to a user is the issue of correctly determining the user's intent. The Clavin project is to build an intelligent natural language query information management system. Clavin must maintain a dynamic user model of the relevant concepts in the user inquiries as they relate to the information sources. The primary goal of Clavin is to autonomously react to changes in user intent as well as the information sources, by dynamically constructing the appropriate queries relative to the changes identified. In this paper, we discuss the... | [
2738
] | Test |
684 | 2 | RF*IPF: A Weighting Scheme for Multimedia Information Retrieval Region-based approach has become a popular research trend in the field of multimedia database retrieval. In this paper, we present the Region Frequency and Inverse Picture Frequency (RF*IPF) weighting, a measure developed to unify region-based multimedia retrieval systems with textbased information retrieval systems. The weighting measure gives the highest weight to regions that occur often in a small number of images in the database. These regions are considered discriminators. With this weighting measure, we can blend image retrieval techniques with TF*IDFbased text retrieval techniques for large-scale Web applications. The RF*IPF weighting has been implemented as a part of our experimental SIMPLIcity image retrieval system and tested on a database of about 200,000 general-purpose images. Experiments have shown that this technique is effective in discriminating images of different semantics. Additionally, the overall similarity approach enables a simple querying interface for multimedia information retrieval systems. 1 | [
2891
] | Train |
685 | 0 | MAP: Design and Implementation of a Mobile Agents Platform The recent development of telecommunication networks has contributed to the success of applications such as information retrieval and electronic commerce, as well as all the services that take advantage of communication in distributed systems. In this area, the emerging technology of mobile agents aroused considerable interest. Mobile agents are applications that can move through the network for carrying out a given task on behalf of the user. In this work we present a platform (called MAP (Mobile Agents Platform)) for the development and the management of mobile agents. The language used both for developing the platform and for carrying out the agents is Java. The platform gives the user all the basic tools needed for creating some applications based on the use of agents. It enables us to create, run, suspend, resume, deactivate, reactivate local agents, to stop their execution, to make them communicate each other and migrate. Keywords: mobile agents, distributed computing, Java, net... | [
561,
613,
807,
2773
] | Test |
686 | 0 | Klava: a Java Framework for Distributed and Mobile Applications Highly distributed networks have now become a common infrastructure for a new kind of wide-area distributed applications whose key design principle is network awareness, namely the ability of dealing with dynamic changes of the network environment. Network-aware computing has called for new programming languages that exploit the mobility paradigm as a basic interaction mechanism. In this paper we present the architecture of Klava, an experimental Java framework for distributed applications and code mobility. We explain how Klava implements code mobility by relying on Java and show a few distributed applications that exploit mobile code and are programmed in Klava. Keywords: Code Mobility, Distributed Applications, Network Awareness, Language and Middleware Implementation, Tuple Spaces, Java. 1 | [
580,
1457,
2611
] | Train |
687 | 3 | Optimizing Large Join Queries in Mediation Systems In data integration systems, queries posed to a mediator need to be translated into a sequence of queries to the underlying data sources. In a heterogeneous environment, with sources of diverse and limited query capabilities, not all the translations are feasible. In this paper, we study the problem of finding feasible and efficient query plans for mediator systems. We consider conjunctive queries on mediators and model the source capabilities through attribute-binding adornments. We use a simple cost model that focuses on the major costs in mediation systems, those involved with sending queries to sources and getting answers back. Under this metric, we develop two algorithms for source query sequencing -- one based on a simple greedy strategy and another based on a partitioning scheme. The first algorithm produces optimal plans in some scenarios, and we show a linear bound on its worst case performance when it misses optimal plans. The second algorithm generates optimal plans in more scenarios, while having no bound on the margin by which it misses the optimal plans. We also report on the results of the experiments that study the performance of the two algorithms. | [
923,
3062
] | Train |
688 | 1 | A Comparative Evaluation of Meta-Learning Strategies over Large and Distributed Data Sets There has been considerable interest recently in various approaches to scaling up machine learning systems to large and distributed data sets. We have been studying approaches based upon the parallel application of multiple learning programs at distributed sites, followed by a meta-learning stage to combine the multiple models in a principled fashion. In this paper, we empirically determine the "best" data partitioning scheme for a selected data set to compose "appropriatelysized " subsets and we evaluate and compare three di#erent strategies, Voting, Stacking and Stacking with Correspondence Analysis (SCANN) for combining classification models trained over these subsets. We seek to find ways to e#ciently scale up to large data sets while maintaining or improving predictive performance measured by the error rate, a cost model, and the TP-FP spread. Keywords: classification, multiple models, meta-learning, stacking, voting, correspondence analysis, data partitioning Email address of co... | [
934,
2242
] | Validation |
689 | 3 | Linearly Bounded Reformulations of Conjunctive Databases (Extended Abstract) Database reformulation is the process of rewriting the data and rules of a deductive database in a functionally equivalent manner. | [
40,
1612
] | Train |
690 | 1 | A Bayesian Computer Vision System for Modeling Human Interactions AbstractÐWe describe a real-time computer vision and machine learning system for modeling and recognizing human behaviors in a visual surveillance task [1]. The system is particularly concerned with detecting when interactions between people occur and classifying the type of interaction. Examples of interesting interaction behaviors include following another person, altering one's path to meet another, and so forth. Our system combines top-down with bottom-up information in a closed feedback loop, with both components employing a statistical Bayesian approach [2]. We propose and compare two different state-based learning architectures, namely, HMMs and CHMMs for modeling behaviors and interactions. The CHMM model is shown to work much more efficiently and accurately. Finally, to deal with the problem of limited training data, a synthetic ªAlife-styleº training system is used to develop flexible prior models for recognizing human interactions. We demonstrate the ability to use these a priori models to accurately classify real human behaviors and interactions with no additional tuning or training. Index TermsÐVisual surveillance, people detection, tracking, human behavior recognition, Hidden Markov Models. 1 | [
51,
919,
2107,
2316,
2906
] | Test |
691 | 4 | Exploring Brick-Based Camera Control Introduction BUILD-IT is a planning tool based on computer vision technology, with a capacity for complex planning and composition tasks (Rauterberg et al. 1997). The system enables users, grouped around a table, to interact in a virtual scene, using real bricks to select and manipulate objects in the scene (Fig. 1, left). A plan view of the scene is projected onto the table. A perspective view of the scene, called side view, is projected on the wall. The plan view contains a storage space with originals, allowing users to select new objects. Object selection is done by putting a brick at the object position. Once selected, objects can be positioned, rotated and fixed by simple brick manipulation (Fig. 1, right). They are de-selected, and stay put, when the brick is covered or removed. Objects brought back to the storage space are deleted. Figure 1: BUILD-IT offers a plan view for combined action and perception, and a<F1 | [
1938
] | Test |
692 | 4 | Implicit Human Computer Interaction Through Context this paper we concentrate on the last case, knowing that in most scenarios a combination of all four cases is the way of choice | [
1027,
1825,
3087
] | Train |
693 | 5 | Probabilistic Roadmap Methods are Embarrassingly Parallel In this paper we report on our experience parallelizing probabilistic roadmap motion planning methods (prms). We show that significant, scalable speedups can be obtained with relatively little effort on the part of the developer. Our experience is not limited to prms, however. In particular, we outline general techniques for parallelizing types of computations commonly performed in motion planning algorithms, and identify potential difficulties that might be faced in other efforts to parallelize sequential motion planning methods. 1 Introduction Automatic motion planning has application in many areas such as robotics, virtual reality systems, and computer-aided design. Although many different motion planning methods have been proposed, most are not used in practice since they are computationally infeasible except for some restricted cases, e.g., when the robot has very few degrees of freedom (dof) [12, 16]. Indeed, there is strong evidence that any complete planner (one that is guaran... | [
841,
1003
] | Train |
694 | 1 | A Neural Network Model for Monotonic Diagnostic Problem Solving The task of diagnosis is to find a hypothesis that best explains a set of manifestations (observations). Generally, it is computationally expensive to find a hypothesis because the number of the potential hypotheses is exponentially large. Recently, many efforts have been made to find parallel processing methods to solve the above difficulty. In this paper, we propose a neural network model for diagnostic problem solving where a diagnostic problem is treated as a combinatorial optimisation problem. One feature of the model is that the causal network is directly used as the network. Another feature is that the errors between the observations and the current activations of manifestation nodes are used to guide the network computing for finding optimal diagnostic hypotheses. 1 Introduction For a set of manifestations(observations), the diagnostic inference is to find the most plausible faults or disorders which can explain why the manifestations are present. In general, an individual d... | [
2411
] | Test |
695 | 0 | Secure Object Spaces - A coordination model for Agents Coordination languages based on shared data spaces are well suited to programming parallel applications composed of cooperating software components. Secure Object Spaces (SOS) extend this model to support composition of independently developed, mutually suspicious, software components. SOS provides flexible and efficient security facilities based on a cryptographic re-interpretation of object types. We discuss the implementation of secure object spaces in the context of a Java-based mobile agent system. 1 Introduction Coordination languages based on shared data spaces have been around for over fifteen years. Oftentimes, researchers have advocated their use for structuring distributed and concurrent systems. This because the mode of communication provided by these languages, sometimes called generative communication, is anonymous --- processes interact by reading and writing entries in a shared space without having to know their interlocutor(s); and is uncoupled --- processes are not ... | [
1902
] | Train |
696 | 4 | Speech/Gesture Interface to a Visual Computing Environment for Molecular Biologists In recent years there has been tremendous progress in 3D, immersive display and virtual reality (VR) technologies. Scientific visualization of data is one of many applications that has benefited from this progress. To fully exploit the potential of these applications in the new environment there is a need for "natural" interfaces that allow the manipulation of such displays without burdensome attachments. This paper describes the use of visual hand gesture analysis enhanced with speech recognition for developing a bimodal gesture/speech interface for controlling a 3-D display. The interface augments an existing application, VMD, which is a VR visual computing environment for molecular biologists. The free hand gestures are used for manipulating the 3-D graphical display together with a set of speech commands. We concentrate on the visual gesture analysis techniques used in developing this interface. The dual modality of gesture/speech is found to greatly aid the interaction capability.... | [
1454
] | Train |
697 | 2 | Accessing Information and Services on the DAML-Enabled Web The DARPA Agent Markup Language (DAML) program aims to allow one to mark up web pages to indicate the meaning of their content; it is intended that the results delivered by a DAML-enabled browser will more closely match the intentions of the user than is possible with today's syntactically oriented search engines. In this paper we present our vision of a DAML-enabled search architecture. We present a set of queries of increasing complexity that should be answered efficiently in a Semantic Web. We describe several scenarios illustrating how queries are processed, identifying the main software components necessary to facilitate the search. We examine the issue of inference in search, and we address how to characterize procedures and services in DAML, enabling a DAML query language to find web sites with specified capabilities. Key Words: Semantic Web, DAML, inference, Web services, process modeling. 1. | [
78,
2666
] | Train |
698 | 3 | Two Algebraic Process Semantics for Contextual Nets Abstract We show that the so-called ‘Petri nets are monoids ’ approach initiated by Meseguer and Montanari can be extended from ordinary place/transition Petri nets to contextual nets by considering suitable nonfree monoids of places. The algebraic characterizations of net concurrent computations we provide cover both the collective and the individual token philosophy, uniformly along the two interpretations, and coincide with the classical proposals for place/transition Petri nets in the absence of read-arcs. | [
733
] | Test |
699 | 0 | Bilateral Negotiation with Incomplete and Uncertain Information: a Decision-Theoretic Approach using a Model of the Opponent . The application of software agents to e-commerce has made a radical change in the way businesses and consumer to consumer transactions take place. Agent negotiation is an important aspect of e-commerce to bring satisfactory agreement in business transactions. We approach e-commerce and negotiation in the context of a distributed multiagent peer help system, I-Help, supporting students in a university course. Personal agents keep models of student preferences and negotiate on their behalf to acquire resources (help) from other agents. We model negotiation among personal agents by means of influence diagram, a decision theoretic tool. To cope with the uncertainty inherent in a dynamic market with self-interested participants, the agents create models of their opponents during negotiation, which help them predict better their opponents' actions. We carried out experiments comparing the proposed negotiation mechanism with influence diagram, one using in addition a model of t... | [
2173,
2584,
2883
] | Validation |
700 | 5 | An On-Line EM Algorithm Applied to Kernel PCA Kernel principal component analysis (Pca) is a recent method for non-linear feature extraction. Applying kernel Pca to a data set with N patterns requires storing and nding the eigenvectors of a N N kernel matrix. This paper describes how an Expectation-Maximization (Em) algorithm for standard Pca can be adapted to kernel Pca without having to store the kernel matrix. Experimental results are given where Em for kernel Pca extracts up to 512 non-linear features from a data set with 15,000 examples. The extracted features lead to good performance when used as pre-processed data for a linear classifier. A novel on-line Em algorithm for Pca is presented and shown to further speed up the learning process. | [
1673
] | Train |
701 | 0 | Virtual Enterprise Design - BDI Agents vs. Objects Current research identifying architectures for a virtual enterprise has moved from information modelling to role modelling. Thus, a high level of autonomy results from the distribution of responsibilities, capabilities, and knowledge among different business units in the virtual enterprise at the design stage. Current trends tend towards using object-oriented technology as an effective abstract system design and implementation methodology. We argue that applying the software agent paradigm to the virtual enterprise provides various advantages on both the design and operational levels. We further show that the Belief Desire Intention agent architecture has additional abilities of mapping real world business unit autonomy and interaction. We also introduce the Belief Desire Intention agent paradigm capability of facilitating highly flexible (agile) enterprise design and implementation. 1 | [
2364,
3177
] | Test |
702 | 2 | Discovering Unexpected Information from Your Competitors' Web Sites Ever since the beginning of the Web, finding useful information from the Web has been an important problem. Existing approaches include keyword-based search, wrapper-based information extraction, Web query and user preferences. These approaches essentially find information that matches the user's explicit specifications. This paper argues that this is insufficient. There is another type of information that is also of great interest, i.e., unexpected information, which is unanticipated by the user. Finding unexpected information is useful in many applications. For example, it is useful for a company to find unexpected information about its competitors, e.g., unexpected services and products that its competitors offer. With this information, the company can learn from its competitors and/or design counter measures to improve its competitiveness. Since the number of pages of a typical commercial site is very large and there are also many relevant sites (competitors), it is very difficult for a human user to view each page to discover the unexpected information. Automated assistance is needed. In this paper, we propose a number of methods to help the user find various types of unexpected information from his/her competitors' Web sites. Experiment results show that these techniques are very useful in practice and also efficient. Keywords Information interestingness, Web comparison, Web mining. 1. | [
471,
815,
1201,
2503,
2864
] | Train |
703 | 2 | Named Entity Recognition from Diverse Text Types Current research in Information Extraction tends to be focused on application-specific systems tailored to a particular domain. The Muse system is a multi-purpose Named Entity recognition system which aims to reduce the need for costly and time-consuming adaptation of systems to new applications, with its capability for processing texts from widely di#ering domains and genres. Although the system is still under development, preliminary results are encouraging, showing little degradation when processing texts of lower quality or of unusual types. The system currently averages 93% precision and 95% recall across a variety of text types. | [
1295
] | Validation |
704 | 0 | Enterprise Agents Solving Conflicts: The cobac Approach . Autonomous agents seem to be a promising approach for application in computermediated supply webs. Supporting the management and integration of the planning, scheduling, and controlling processes they can be used as "enterprise delegates". However, this leads to a problem with common autonomous agent architectures, as they are not designed to model complex decision behavior of entire companies. In this paper we present an innovative approach to model Intelligent Agents in the context of the new agent architecture "Enterprise Agents", which is using conflicts of interests explicitly to determine agent's behavior and discuss its application to eBusiness. Keywords. Agent Control, Conflict Management, Adaptation, BDI-Architecture, Enterprise Agents 1. | [
3167
] | Test |
705 | 4 | Evaluation of Recommender Algorithms for an Internet Information Broker Based Association rules are a widely used technique to generate recommendations in commercial and research recommender systems. Since more and more Web sites, especially of retailers, offer automatic recommender services using Web usage mining, evaluation of recommender algorithms becomes increasingly important. In this paper we first present a framework for the evaluation of different aspects of recommender systems based on the process of discovering knowledge in databases of Fayyad et al. and then we focus on the comparison of the performance of two recommender algorithms based on frequent itemsets. The first recommender algorithm uses association rules, and the other recommender algorithm is based on the repeat-buying theory known from marketing research. For the evaluation we concentrated on how well the patterns extracted from usage data match the concept of useful recommendations of users. We use 6 month of usage data from an educational Internet information broker and compare useful recommendations identified by users from the target group of the broker with the results of the recommender algorithms. The results of the evaluation presented in this paper suggest that frequent itemsets from purchase histories match the concept of useful recommendations expressed by users with satisfactory accuracy (higher than 70%) and precision (between 60% and 90%). Also the evaluation suggests that both algorithms studied in the paper perform similar on real-world data if they are tuned properly. | [
2250
] | Train |
706 | 4 | End of First Year Report displays of information (graphs, plots, etc.) are a recent invention at around 1750-1800 [15]. Andrews defines Information Visualisation as the visual presentation of information spaces and structures to facilitate their rapid assimilation and understanding [149]. In the same document, the authors give a collection of Information Visualisation pointers (references also available at http://www.iicm.edu/hci/ivis). A more complete on-line document for Information Visualisation, from Andrews is available at http://www.iicm.edu.hci/ivis/node2.htm. A report on three-dimensional Information Visualisation is given by Young, also available on-line, at http://www.dur.ac.uk/~dcs3py/pages/work/documents/litsurvey /IV-Survey/ [150]. This report gives a visualisation techniques enumeration and a survey of research visualisation systems. Two other Web resources for Information Visualisation are Olive (Online of Information Visualisation Environments - http://otal.umd.edu/Olive/), and the CS348 course... | [
1017
] | Validation |
707 | 2 | Self-Supervised Learning for Visual Tracking and Recognition of Human Hand Due to the large variation and richness of visual inputs, statistical learning gets more and more concerned in the practice of visual processing such as visual tracking and recognition. Statistical models can be trained from a large set of training data. However, in many cases, since it is not trivial to obtain a large labeled and representative training data set, it would be difficult to obtain a satisfactory generalization. Another difficulty is how to automatically select good features for representation. By combining both labeled and unlabeled training data, this paper proposes a new learning paradigm, selfsupervised learning, to investigate the issues of learning bootstrapping and model transduction. Inductive learning and transductive learning are the two main cases of self-supervised learning, in which the proposed algorithm, Discriminant-EM (D-EM), is a specific learning technique. Vision-based gesture... | [
506,
2808
] | Validation |
708 | 4 | Ambient Displays: Turning Architectural Space into an Interface between People and Digital Information . We envision that the architectural space we inhabit will be a new form of interface between humans and online digital information. This paper discusses Ambient Displays: a new approach to interfacing people with online digital information. Ambient Displays present information within a space through subtle changes in light, sound, and movement, which can be processed in the background of awareness. We describe the design and implementation of two example Ambient Displays, the ambientROOM and Ambient Fixtures. Additionally, we discuss applications of Ambient Displays and propose theories of design of such interfaces based on our initial experiences. 1 INTRODUCTION Ambient \Am"bi*ent\, a. Surrounding, encircling, encompassing, and environing. -Oxford English Dictionary Display \Dis*play"\, n. An opening or unfolding; exhibition; manifestation. -Webster's Revised Unabridged Dictionary (1913) Nature is filled with subtle, beautiful and expressive ambient displays that engage ... | [
1304,
1619
] | Train |
709 | 4 | Gathering User Interface Design Requirements for Social Computing Design for cooperation is a challenge. As designers we note that as we are moving towards the final years of this century, several areas have achieved significant breakthroughs. Among them, it is easy to perceive that areas of Computing and Telecommunications have had an impact of paramount importance to society as a whole. These technologies have allowed an increasing integration of research fields, people of various backgrounds and abilities as well as made the interaction of different cultures possible. As a result, we have been living in the Internet era with a very large number of Web sites which can be visited, queried and played with. That constitutes what we call social computing. Application examples are: digital libraries, health care information systems, Physics collaboratories, and Web-based entertainments like interactive Web games. Within this context, we are concerned with the user interface design requirements gathering for such systems. In that sense, we present a prot... | [
1742
] | Train |
710 | 2 | Building a Machine Learning Based Text Understanding System Text understanding systems are approaching the point of being a practical technology as long as the system is trained for a narrowly defined domain. Machine learning and statistical approaches can minimize the effort involved in adapting a text understanding system to a new domain. This paper presents a system whose goal is deep understanding, limited only by the necessity of designing a formal representation of the target concepts relevant to the domain. This system is an advance over previous machine learning based systems because of its richer output representation, and an advance over equally expressive text understanding systems because of its more extensive use of machine learning. 1 Information Extraction from Free Text A variety of systems have been developed in recent years that extract information from text. None of them attempts general-purpose understanding, but instead focus on narrowly defined information needs. A domain is defined as a collection of docum... | [
47,
2068,
2961
] | Validation |
711 | 3 | On Caching Search Engine Results In this paper we explore the problem of Caching of Search Engine Query Results in order to reduce the computing and I/O requirements needed to support the functionality of a search engine of the world-wide web. Based on traces from search engines we show that there is significant locality in the queries asked, that is, 20-30% of the queries have been previously submitted by the same or a different user. Using trace-driven simulation we show that medium-sized caches can hold most of the frequently-submitted queries. Finally, we propose and evaluate a new cache replacement algorithm named LRU-2S, that takes into account both the frequency and the recency of access to a page when making a replacement decision. | [
1050,
1376,
2609
] | Validation |
712 | 5 | A Tableau-Based Proof Method for Temporal Logics of Knowledge and Belief . In this paper we define two logics, KLn and BLn , and present tableau-based decision procedures for both. KLn is a temporal logic of knowledge. Thus, in addition to the usual connectives of linear discrete temporal logic, it contains a set of unary modal connectives for representing the knowledge possessed by agents. The logic BLn is somewhat similar; it is a temporal logic that contains connectives for representing the beliefs of agents. In addition to a complete formal definition of the two logics and their decision procedures, the paper includes a brief review of their applications in AI and mainstream computer science, correctness proofs for the decision procedures, a number of worked examples illustrating the decision procedures, and some pointers to further work. KEYWORDS:Temporal logics of knowledge and belief, theorem proving, tableau. 1 Introduction This paper presents two logics, called KLn and BLn respectively, and gives tableau-based decision procedures for both. The l... | [
108,
183,
353,
2059,
2338
] | Validation |
713 | 1 | GIB: Steps Toward an Expert-Level Bridge-Playing Program This paper describes Goren In a Box (gib), the first bridge-playing program to approach the level of a human expert. We give a basic overview of the algorithms used, describe their strengths and weaknesses, and present the results of experiments comparing gib to both human opponents and earlier programs. Introduction Of all the classic games of skill, only card games and Go have yet to see the appearance of serious computer challengers. In Go, this appears to be because the game is fundamentally one of pattern recognition as opposed to search; the brute-force techniques that have been so successful in the development of chess-playing programs have failed almost utterly to deal with Go's huge branching factor. Indeed, the arguably strongest Go program in the world was beaten by Janice Kim in the AAAI-97 Hall of Champions after Kim had given the program a monumental 25 stone handicap. Card games appear to be different. Perhaps because they are games of imperfect information, or perhaps... | [
1806
] | Train |
714 | 1 | Active Learning for Natural Language Parsing and Information Extraction In natural language acquisition, it is difficult to gather the annotated data needed for supervised learning; however, unannotated data is fairly plentiful. Active learning methods attempt to select for annotation and training only the most informative examples, and therefore are potentially very useful in natural language applications. However, existing results for active learning have only considered standard classification tasks. To reduce annotation effort while maintaining accuracy, we apply active learning to two non-classification tasks in natural language processing: semantic parsing and information extraction. We show that active learning can significantly reduce the number of annotated examples required to achieve a given level of performance for these complex tasks. Keywords: active learning, natural language learning, learning for parsing, learning for information extraction Email address of contact author: cthomp@csli.stanford.edu Phone number of contact author: (650)8... | [
371,
875,
1928,
2068,
2466,
2961
] | Validation |
715 | 5 | SNePS: A Logic for Natural Language Understanding and Commonsense Reasoning The use of logic for knowledge representation and reasoning systems is controversial. There are, indeed, several ways that standard First Order Predicate Logic is inappropriate for modelling natural language understanding and commonsense reasoning. However, a more appropriate logic can be designed. This chapter presents several aspects of such a logic. 1 Introduction My colleagues, students, and I have been engaged in a long-term project to build a natural language using intelligent agent. While our approach to natural language understanding (NLU) and commonsense reasoning (CSR) has been logic-based, we have thought that the logics developed for metamathematics, e.g. (Kleene, 1950), are not the best ones for our purpose. Instead, we have designed new logics, better suited for NLU and CSR. The current version of these logics constitutes the formal language and inference mechanism of the knowledge representation/reasoning (KRR) system, SNePS 2.4 (Shapiro and The SNePS Implementation Gr... | [
1292
] | Train |
716 | 5 | Diagnosis and Decision Making in Normative Reasoning . Diagnosis theory reasons about incomplete knowledge and only considers the past. It distinguishes between violations and non-violations. Qualitative decision theory reasons about decision variables and considers the future. It distinguishes between fulfilled goals and unfulfilled goals. In this paper we formalize normative diagnoses and decisions in the special purpose formalism DIO(DE) 2 as well as in extensions of the preference-based deontic logic PDL. The DIagnostic and DEcision-theoretic framework for DEontic reasoning DIO(DE) 2 formalizes reasoning about violations and fulfillments, and is used to characterize the distinction between normative diagnosis theory and (qualitative) decision theory. The extension of the preference-based deontic logic PDL shows how normative diagnostic and decision-theoretic reasoning -- i.e. reasoning about violations and fulfillments -- can be formalized as an extension of deontic reasoning. 1. Introduction In the AI and Law literature it is... | [
1459,
1658,
2209
] | Train |
717 | 0 | Autonomous Agents For Business Process Management : Traditional approaches to managing business processes are often inadequate for large-scale, organisation -wide, dynamic settings. However since Internet and Intranet technologies have become widespread, an increasing number of business processes exhibit these properties. Therefore a new approach is needed. To this end, we describe the motivation, conceptualisation, design and implementation of a novel agent-based business process management system. The key advance of our system is that responsibility for enacting various components of the business process is delegated to a number of autonomous problem solving agents. To enact their role, these agents typically interact and negotiate with other agents in order to coordinate their actions and to buy in the services they require. This approach leads to a system that is significantly more agile and robust than its traditional counterparts. To help demonstrate these benefits, a companion paper describes the application of our system to a ... | [
1606,
1724,
2344
] | Validation |
718 | 4 | Real-Time Error in Location Modeling for Ubiquitous Computing No matter which technologies or techniques a ubiquitous location system uses, its measurements will have some amount of quantifiable error. | [
2426,
3159
] | Train |
719 | 1 | Learning the Face Space - Representation and Recognition This paper advances an integrated learning and evolutionary computation methodology for approaching the task of learning the face space. The methodology is geared to provide a framework whereby enhanced and robust face coding and classification schemes can be derived and evaluated using both machine and human benchmark studies. In particular we take an interdisciplinary approach, drawing from the accumulated and vast knowledge of both the computer vision and psychology communities, and describe how evolutionary computation and statistical learning can engage in mutually beneficial relationships in order to define an exemplar (Absolute)-Based Coding (ABC) multidimensional face space representation for successfully coping with changing population (face) types, and to leverage past experience for incremental face space definition. 1. Introduction Among the most challenging tasks for visual form (`shape') analysis and object recognition are understanding how people process and recognize ... | [
2328,
2751
] | Train |
720 | 3 | First-order expressivity for S5-models: modal vs. two-sorted languages this paper we are going to prove some results on the expressive power of the standard first-order modal language by comparing it with its extensional counterpart. We thereby restrict our attention to the case where the modal language is interpreted on S5-models. Moreover, we decided to deal exclusively with constant domain models, that is with models in which the domains of all worlds are the same. It is worth mentioning, however, that our method can be applied to logics based on varying domain models as well. Before we describe the results of this paper in more detail, we hasten to add that there exists some work done by other authors to which our results are related. In [8, 9, 10] Fine proves, among other things, a number of preservation results for modal first-order formulas which are relevant for certain philosophical distinctions. For instance, he provides a semantical characterization of de dicto formulas (within S5) and investigates conditions under which de re formulas are eliminable in certain extensions of S5 (see also [6, 18]). In [3] it was shown 2 that a remarkable portion of classical model theory can be transferred to the domain of modal logic. More closely related to our work are [14] and [15]. They discuss a number of formulas from the two-sorted language that are not expressible in the modal language. | [
262
] | Train |
721 | 1 | Web Mining in Soft Computing Framework: Relevance, State of the Art and Future Directions This paper summarizes the different characteristics of web data, the basic components of web mining and its different types, and their current states of the art. The reason for considering web mining, a separate field from data mining, is explained. The limitations of some of the existing web mining methods and tools are enunciated, and the significance of soft computing (comprising fuzzy logic (FL), artificial neural networks (ANNs), genetic algorithms (GAs), and rough sets (RSs) highlighted. A survey of the existing literature on "soft web mining" is provided along with the commercially available systems. The prospective areas of web mining where the application of soft computing needs immediate attention are outlined with justification. Scope for future research in developing "soft web mining" systems is explained. An extensive bibliography is also provided. | [
255,
605,
759,
901,
921,
1445,
1678,
2068,
2503,
2532,
2918
] | Validation |
722 | 1 | Active Perception and Map Learning for Robot Navigation This paper describes a simulated on-line mapping system for robot navigation. This system allows the autonomous creation of topological maps enhanced with metrical information provided by internal (odometry) and external (vision and sonars) sensors. Within such maps, the robot's position is represented and calculated probabilistically according to algorithms that are inspired by Hidden Markov Models. The visual system is very simple and does not allow reliable recognition of speci c places but, used jointly with odometry, sonar recordings and an active perception system, it allows reliable localization even when the robot starts exploring its environment, and when it is passively translated from one place to another. Advantages and drawbacks of the current system are discussed, together with means to remediate the latter. | [
588
] | Train |
723 | 1 | Regularizing AdaBoost Boosting methods maximize a hard classification margin and are known as powerful techniques that do not exhibit overfitting for low noise cases. Also for noisy data boosting will try to enforce a hard margin and thereby give too much weight to outliers, which then leads to the dilemma of non-smooth fits and overfitting. Therefore we propose three algorithms to allow for soft margin classification by introducing regularization with slack variables into the boosting concept: (1) AdaBoost reg and regularized versions of (2) linear and (3) quadratic programming AdaBoost. Experiments show the usefulness of the proposed algorithms in comparison to another soft margin classifier: the support vector machine. 1 Introduction Boosting and other ensemble methods have been used with success in several applications, e. g. OCR [12, 7]. For low noise cases several lines of explanation have been proposed as candidates for explaining the well functioning of boosting methods. (a) Breiman proposed that ... | [
191,
1299
] | Test |
724 | 1 | The Case Against Accuracy Estimation for Comparing Induction Algorithms We analyze critically the use of classification accuracy to compare classifiers on natural data sets, providing a thorough investigation using ROC analysis, standard machine learning algorithms, and standard benchmark data sets. The results raise serious concerns about the use of accuracy for comparing classifiers and draw into question the conclusions that can be drawn from such studies. In the course of the presentation, we describe and demonstrate what we believe to be the proper use of ROC analysis for comparative studies in machine learning research. We argue that this methodology is preferable both for making practical choices and for drawing scientific conclusions. 1 INTRODUCTION Substantial research has been devoted to the development and analysis of algorithms for building classifiers, and a necessary part of this research involves comparing induction algorithms. A common methodology for such evaluations is to perform statistical comparisons of the accuracies of learned class... | [
259,
744,
934,
2513,
2728
] | Validation |
725 | 0 | Integrating Peer-to-Peer Networking and Computing in the Agentscape Framework The combination of peer-to-peer networking and agentbased computing seems to be a perfect match. Agents are cooperative and communication oriented, while peerto -peer networks typically support distributed systems in which all nodes have equal roles and responsibilities. AgentScape is a framework designed to support large-scale multi-agent systems. Pole extends this framework with peerto -peer computing. This combination facilitates the development and deployment of new agent-based peer-to-peer applications and services. | [
1079,
1738,
2593,
2843,
3089
] | Train |
726 | 0 | Experimental Results with Real-Time Scheduling Using DECAF DECAF [6] is a software toolkit for the rapid design, development, and execution of "intelligent" agents to achieve solutions in complex software systems. From a research community perspective, DECAF provides a modular platform for evaluating and disseminating results in agent architectures, including communication, planning, scheduling, execution monitoring, coordination, diagnosis, and learning. This paper will describe a methodology and results for building and evaluating execution schedules of agent actions in the DECAF architecture. A brief description of DECAF is provided describing how the modular design allows for testing of multiple scheduling algorithms. A description of the types of agents that were used to demonstrate three particular capabilities of DECAF; scalability, parallelism and the threaded nature of the architecture. Lastly, experiments using different scheduling algorithms were utilized to develop an experimental platform for future research in agent scheduling. F... | [
2321
] | Train |
727 | 4 | Real-Time American Sign Language Recognition Using Desk and Wearable Computer Based Video Hidden Markov models (HMM's) have been used prominently and successfully in speech recognition and, more recently, in handwriting recognition. Consequently, they seem ideal for visual recognition of complex, structured hand gestures such as are found in sign language. We describe two experiments that demonstrate a realtime HMM-based system for recognizing sentence level American Sign Language (ASL) without explicitly modeling the fingers. The first experiment tracks hands wearing colored gloves and attains a word accuracy of 99%. The second experiment tracks hands without gloves and attains a word accuracy of 92%. Both experiments have a 40 word lexicon. 1 Introduction While there are many different types of gestures, the most structured sets belong to the sign languages. In sign language, each gesture already has assigned meaning, and strong rules of context and grammar may be applied to make recognition tractable. To date, most work on sign language recognition has employed expensi... | [
120,
664,
765,
1110,
1464,
1556,
1854,
1969,
2119,
2203,
2562,
2799
] | Test |
728 | 2 | A Conceptual Annotation Approach to Indexing in a Web-Based Information System All the specialists have agreed that the possibility of adding to multimedia WWW objects some sort of `conceptual ' annotations describing their information content would greatly contribute to solve the problem of their `intelligent ' indexing and retrieval. We propose to associate with the Web objects not the final conceptual annotation, but a simple natural language (NL) caption, in the form of short texts representing a general, neutral description of their informational content. The NL caption will then be converted into a conceptual annotation in NKRL (Narrative Knowledge Representation Language), making use of an automatic translation system like those we have implemented in the context of recent European projects. 1 Introduction Many applications enabled by the WWW, such as distance learning, electronic commerce, information gathering and filtering, have strong need for tools supporting the effective retrieval of information. It is today well recognized that an effective retriev... | [
2665
] | Train |
729 | 4 | Supporting Cross-Cultural Communication with a Large-Screen System As opportunities for international collaboration and crosscultural communication among people from heterogeneous cultures increase, the importance of electronic communication support is increasing. To support cross-cultural communication, we believe it is necessary to offer environments in which participants enjoy conversations, which allow them to share one anothers background and profile visually. We believe that the following three functions are important: (1) showing topics based on participants profiles and cultural background; (2) life-sized, large-screen interface; and, (3) displaying objects which show feelings of identify. In this paper, we discuss the implementation and the empirical evaluation of two systems that were designed to support cross-cultural communication in the real world or between remote locations. From the empirical evaluation of these systems, we conclude that these systems add new functionality to support conversation contents, which may be especially useful in a cross-cultural context where language skills are an issue, and this type of environment may be especially useful in a pre-collaboration context. | [
200
] | Validation |
730 | 3 | Structure and Performance of Decision Support Algorithms on Active Disks Growth and usage trends for large decision support databases indicate that there is a need for architectures that scale the processing power as the dataset grows. These trends indicate that the processing demand for large decision support databases is growing faster than the improvement in performance of commodity processors. To meet this need, several researchers have recently proposed Active Disk/IDISK architectures which integrate substantial processing power and memory into disk units. In this paper, we examine the utility of Active Disks for decision support databases. We try to answer the following questions. First, is it possible to restructure algorithms for common decision support tasks to utilize Active Disks? Second, how does the performance of Active Disks compare with that of traditional servers for these tasks? Finally, how would Active Disks be integrated into the software architecture of decision support databases? 1 Introduction Growth and usage trends for large decis... | [
3068
] | Train |
731 | 3 | Algorithms for Temporal Query Operators in XML Databases The contents of an XML database or XML/Web data warehouse is seldom static. New documents are created, documents are deleted, and more important: documents are updated. In many cases, we want to be able to search in historical versions, retrieve documents valid at a certain time, query changes to documents, etc. This can be supported by extending the system with temporal database features. In this paper we describe the new query operators needed in order to support an XML query language which supports temporal operations. We also describe the algorithms which can make efficient implementation of these query operators possible. Keywords: XML, temporal databases, query processing 1 | [
416,
1660,
2232,
2524,
2632
] | Validation |
732 | 2 | Document Clustering using Word Clusters via the Information Bottleneck Method We present a novel implementation of the recently introduced information bottleneck method for unsupervised document clustering. Given a joint empirical distribution of words and documents, p(x; y), we first cluster the words, Y , so that the obtained word clusters, Y_hat , maximally preserve the information on the documents. The resulting joint distribution, p(X; Y_hat ), contains most of the original information about the documents, I(X; Y_hat ) ~= I(X;Y ), but it is much less sparse and noisy. Using the same procedure we then cluster the documents, X , so that the information about the word-clusters is preserved. Thus, we first find word-clusters that capture most of the mutual information about the set of documents, and then find document clusters, that preserve the information about the word clusters. We tested this procedure over several document collections based on subsets taken from the standard 20Newsgroups corpus. The results were assessed by calculating the correlation between the document clusters and the correct labels for these documents. Finding from our experiments show that this double clustering procedure, which uses the information bottleneck method, yields significantly superior performance compared to other common document distributional clustering algorithms. Moreover, the double clustering procedure improves all the distributional clustering methods examined here. | [
430,
487,
901,
1726,
2680
] | Train |
733 | 3 | Algebraic Models for Contextual Nets We extend the algebraic approach of Meseguer and Montanari from ordinary place/transition Petri nets to contextual nets, covering both the collective and the individual token philosophy uniformly along the two interpretations of net behaviors. | [
698
] | Validation |
734 | 1 | An Ejection Chain Approach for the Generalized Assignment Problem this paper, we propose an ejection chain approach under the framework of tabu search (TS) for the generalized assignment problem (GAP), which is known to be NP-hard (Sahni and Gonzalez 1976). GAP seeks a minimum cost assignment of n jobs to m agents subject to a resource constraint for each agent. Among various heuristic algorithms developed for GAP are: a combination of the greedy method and local search by Martello and Toth (1981, 1990); a tabu search and simulated annealing approach by Osman (1995); a genetic algorithm by Chu and Beasley (1997); VDS methods by Amini and Racer (1995) and Racer and Amini (1994); a tabu search approach by Laguna et al. (1995) (which is proposed for a generalization of GAP); a set partitioning heuristic by Cattrysse et al. (1994); a relaxation heuristic by Lorena and Narciso (1996); a GRASP and MAX-MIN ant system combined with local search and tabu search by Lourenco and Serra (1998); a linear relaxation heuristic by Trick (1992); and so on. Many exact algorithms have also been proposed (e.g., Nauss 2003, Savelsbergh 1997). A simpler version of an ejection chain approach has also been proposed for the GAP in Laguna et al. (1995). Our ejection chain is based on the idea described in Glover (1997) | [
2729
] | Train |
735 | 1 | On Mapping Decision Trees and Neural Networks There exist several methods for transforming decision trees to neural networks. These methods typically construct the networks by directly mapping decision nodes or rules to the neural units. As a result, the networks constructed are often larger than necessary. This paper describes a pruning-based method for mapping decision trees to neural networks, which can compress the network by removing unimportant and redundant units and connections. In addition, equivalent decision trees extracted from the pruned networks are simpler than those induced by well-known algorithms such as ID3 and C4.5. Keywords: decision trees, neural networks, pruning. 1 Introduction Decision trees have been widely used for nonparametric pattern classification tasks which involve several pattern classes and a large number of features. Given an input pattern, the tree classifier performs the comparisons stipulated in each decision node of the tree, and then branches to either the left or the right subtree based on... | [
584,
1267
] | Validation |
736 | 3 | Atomi - Automated Reconstruction Of Topographic Objects From Aerial Images Using Vectorized Map Information The project ATOMI is a co-operation between the Federal Office of Topography (L+T) and ETH Zurich. The aim of ATOMI is to update vector data of road centerlines and building roof outlines from 1:25,000 maps, fitting it to the real landscape, improve the planimetric accuracy to 1m and derive height information (one representative height for each building) with 1-2 m accuracy. This update should be achieved by using image analysis techniques developed at ETH Zurich and digital aerial imagery. The whole procedure should be implemented as a stand-alone software package, able to import and export data as used at L+T. It should be quasi operational, fast, and the most important reliable. We do not aim at full automation (ca. 80% completeness is a plausible target). The paper will present in detail the aims, input data, strategy and general methods used in ATOMI. We will also present an overview of the results achieved up to now, and problems faced in building and road reconstruction. More de... | [
207
] | Validation |
737 | 2 | Web Document Classification based on Hyperlinks and Document Semantics . Besides the basic content, a web document also contains a set of hyperlinks pointing to other related documents. Hyperlinks in a document provide much information about its relation with other web documents. By analyzing hyperlinks in documents, inter-relationship among documents can be identified. In this paper, we will propose an algorithm to classify web documents into subsets based on hyperlinks in documents and their content. Representativedocuments will also be identified in each subset based on the proposed similarity definition With the representative document, searching for related documents can be achieved. 1 Introduction The WWW is growing through a decentralized process, and documents in the web are lack of logical organization. Besides, the enormous number of web documents make the manipulation and further operation on web documents difficult. Although the size of web document set is large, we need not analyze all the web documents as a whole. Web documents can... | [
471,
1360
] | Test |
738 | 5 | Causal Models of Mobile Service Robot Behavior Temporal projection, the process of predicting what will happen when a robot executes its plan, is essential for autonomous service robots to successfully plan their missions. This paper describes a causal model of the behavior exhibited by the mobile robot Rhino when running concurrent reactive plans for performing office delivery jobs. The model represents aspects of robot behavior that cannot be represented by most action models used in AI planning: it represents the temporal structure of continuous control processes, several modes of their interferences, and various kinds of uncertainty. This enhanced expressiveness enables xfrm (McD92; BM94), a robot planning system, to predict, and therefore forestall, various kinds of behavior flaws including missed deadlines whilst exploiting incidental opportunities. The proposed causal model is experimentally validated using the robot and its simulator. Introduction Temporal projection, the process of predicting what will happen when a ro... | [
1025
] | Train |
739 | 2 | Boosting and Rocchio Applied to Text Filtering We discuss two learning algorithms for text filtering: modified Rocchio and a boosting algorithm called AdaBoost. We show how both algorithms can be adapted to maximize any general utility matrix that associates cost (or gain) for each pair of machine prediction and correct label. We first show that AdaBoost significantly outperforms another highly effective text filtering algorithm. We then compare AdaBoost and Rocchio over three large text filtering tasks. Overall both algorithms are comparable and are quite effective. AdaBoost produces better classifiers than Rocchio when the training collection contains a very large number of relevant documents. However, on these tasks, Rocchio runs much faster than AdaBoost. 1 Introduction With the explosion in the amount of information available electronically, information filtering systems that automatically send articles of potential interest to a user are becoming increasingly important. If users indicate their interests to a filtering system... | [
88,
322,
887,
1215,
1440,
1525,
1577,
2255,
2461,
2676,
2690,
3107
] | Train |
740 | 0 | Impact: A Platform for Collaborating Agents twork. The Impact server provides the infrastructure upon which different Impact agents can interact. To avoid a performance bottleneck, multiple copies of the server can be replicated and scattered across the network. Impact agents A set of data objects can be represented in a wide variety of ways. When building an application, we'd like to select a data structure that supports the application operations that are the most frequently executed, the most critical, or both. So, any definition of an agent must support such flexible choice of data structures, and agentization must let us extend arbitrary data representations. In Impact, an agent consists of any body of software code whatsoever, with the associated wrapper. Figure 2 shows such an agent's architecture. The software code. The agent's code consists of two parts: . a set of data structures (or data types) manipulated by the agent. For example, if we are building a database agen | [
61,
1040,
2956,
2972,
3104
] | Test |
741 | 2 | Efficient Web Form Entry on PDAs We propose a design for displaying and manipulating HTML forms on small PDA screens. The form input widgets are not shown until the user is ready to fill them in. At that point, only one widget is shown at a time. The form is summarized on the screen by displaying just the text labels that prompt the user for each widget's information. The challenge of this design is to automatically find the match between each text label in a form, and the input widget for which it is the prompt. We developed eight algorithms for performing such label-widget matches. Some of the algorithms are based on n-gram comparisons, while others are based on common form layout conventions. We applied a combination of these algorithms to 100 simple HTML forms with an average of four input fields per form. These experiments achieved a 95% matching accuracy. We developed a scheme that combines all algorithms into a matching system. This system did well even on complex forms, achieving 80% accuracy in our experiments involving 330 input fields spread over 48 complex forms. | [
1909,
2315
] | Validation |
742 | 4 | Constructing, Organizing, and Visualizing Collections of Topically Related Web Resources For many purposes, the Web page is too small a unit of interaction and analysis. Web sites are structured multimedia documents consisting of many pages, and users often are interested in obtaining and evaluating entire collections of topically related sites. Once such a collection is obtained, users face the challenge of exploring, comprehending, and organizing the items. We report four innovations that address these user needs. • We replaced the web page with the web site as the basic unit of interaction and analysis. • We defined a new information structure, the clan graph, that groups together sets of related sites. • We augment the representation of a site with a site profile, information about site structure and content that helps inform user evaluation of a site. • We invented a new graph visualization, the auditorium visualization, that reveals important structural and content properties of sites within a clan graph. Detailed analysis and user studies document the utility of this approach. The clan graph construction algorithm tends to filter out irrelevant sites and discover additional relevant items. The auditorium visualization, augmented with drill down capabilities to explore site profile data, helps users to find high-quality sites as well as sites that serve a particular function. | [
471,
505,
1983,
2257
] | Train |
743 | 1 | Finding Counterexamples to Inductive Conjectures We present an implementation of a method for £nding counterexamples to universally quanti£ed inductive conjectures in £rst-order logic. Our method uses the proof by consistency strategy to guide a search for a counterexample and a standard £rst-order theorem prover to perform a concurrent check for inconsistency. We explain brie¤y the theory behind the method, describe our implementation, and evaluate results achieved on a variety of incorrect conjectures from various sources. Some work in progress is also presented: we are applying the method to the veri£cation of cryptographic security protocols. In this context, a counterexample to a security property can indicate an attack on the protocol, and our method extracts the trace of messages exchanged in order to effect the attack. This application demonstrates the advantages of the method, in that quite complex side conditions decide whether a particular sequence of messages is possible. Using a theorem prover provides a natural way of dealing with this. Some early results are presented and we discuss future work. 1 | [
486
] | Train |
744 | 1 | A Machine Learning Researcher's Foray into Recidivism Prediction We discuss an application of machine learning to recidivism prediction. Our initial results motivate the need for a methodology for technique selection for applications that involve unequal but unknown error costs, a skewed data set, or both. Evaluation methodologies traditionally used in machine learning are inadequate for analyzing performance in these situations, although they arise frequently when addressing real-world problems. After discussing the problem of recidivism prediction and the particulars of our data set, we present experimental results that motivate the need to evaluate learning algorithm over a range of error costs. We then describe Receiver Operating Characteristic (ROC) analysis, which has been used extensively in signal detection theory for decades but has only recently begun to filter into machine learning research. With this new perspective, we revisit the recidivism prediction task and present results that contradict those obtained using a traditional ... | [
724,
1737
] | Validation |
745 | 1 | Induction of decision trees using RELIEFF In the context of machine learning from examples this paper deals with the problem of estimating the quality of attributes with and without dependencies between them. Greedy search prevents current inductive machine learning algorithms to detect significant dependencies between the attributes. Recently, Kira and Rendell developed the RELIEF algorithm for estimating the quality of attributes that is able to detect dependencies between attributes. We show strong relation between RELIEF's estimates and impurity functions, that are usually used for heuristic guidance of inductive learning algorithms. We propose to use RELIEFF, an extended version of RELIEF, instead of myopic impurity functions. We have reimplemented Assistant, a system for top down induction of decision trees, using RELIEFF as an estimator of attributes at each selection step. The algorithm is tested on several artificial and several real world problems. Results show the advantage of the presented approach to inductive lea... | [
2502
] | Train |
746 | 4 | Situated Computing: A Paradigm to Enhance the Mobile User's Interaction When people interact with computers, they have to pay attention for things that are not related to the situation of the problems because the interfaces are not contextualized to their working environment. Hence it is sometimes hard to integrate computers as embedded tools, which facilitate users to accomplish their objectives easily in the working life. Situated computing is a new paradigm for mobile computer users based on their physical context and activities carried out in the workspace. It defines the infrastructure how the situated interaction can be provided using applications. In this chapter we first describe a model called situation metaphor to design interaction between the user and mobile computers as the basis for the situated computing. Thereafter, a framework called Situated Information Filing and Filtering (SIFF) is presented as the foundation for situated application development. In general a three-stages schema is given considerting the top stage for situated applications. Four applications based on the SIFF are also presented to demonstrate the enhancement of mobile user's interaction that can be achieved. | [
1900
] | Train |
747 | 1 | SFS Based View Synthesis for Robust Face Recognition Sensitivity to variations in pose is a challenging problem in face recognition using appearance-based methods. More specifically, the appearance of a face changes dramatically when viewing and/or lighting directions change. Various approaches have been proposed to solve this difficult problem. They can be broadly divided into three classes: 1) multiple image based methods where multiple images of various poses per person are available, 2) hybrid methods where multiple example images are available during learning but only one database image per person is available during recognition, and 3) single image based methods where no example based learning is carried out. In this paper, we present a method that comes under class 3. This method based on shape-from-shading (SFS) improves the performance of a face recognition system in handling variations due to pose and illumination via image synthesis. 1 Introduction Face recognition has become one of the most active areas of research in image... | [
3058
] | Train |
748 | 1 | On Behavior Classification in Adversarial Environments In order for robotic systems to be successful in domains with other agents possibly interfering with the accomplishing of goals, the agents must be able to adapt to the opponents' behavior. The more quickly the agents can respond to a new situation, the better they will perform. We present an approach to doing adaptation which relies on classification of the current adversary into predefined adversary classes. For feature extraction, we present a windowing technique to abstract useful but not overly complicated features. In order to take into account the spatial locality of topological differences, we use a previously developed similarity metric. The feature extraction and classification steps are fully implemented in the domain of simulated robotic soccer, and experimental results are presented. | [
1526
] | Train |
749 | 1 | Automated State Abstraction for Options using the U-Tree Algorithm Learning a complex task can be significantly facilitated by defining a hierarchy of subtasks. An agent can learn to choose between various temporally abstract actions, each solving an assigned subtask, to accomplish the overall task. In this paper, we study hierarchical learning using the framework of options. We argue that to take full advantage of hierarchical structure, one should perform option-specific state abstraction, and that if this is to scale to larger tasks, state abstraction should be automated. We adapt McCallum's U-Tree algorithm to automatically build option-specific representations of the state feature space, and we illustrate the resulting algorithm using a simple hierarchical task. Results suggest that automated option-specific state abstraction is an attractive approach to making hierarchical learning systems more effective. 1 Introduction Researchers in the field of reinforcement learning have recently focused considerable attention on temporally abst... | [
1964
] | Train |
750 | 0 | A Behaviour-based Approach to Position Selection for Simulated Soccer Agents Selecting an optimal position for each soccer robot to move to in a robot football game is a challenging and complex task since the environment and robot motion are so dynamic and unpredictable. This paper provides an overview of behaviour-based position selection schemes used by Essex Wizards'99 simulated soccer team, a third place in RoboCup'99 simulator league at Stockholm. The focus concentrates on how each position selection behaviour is selected for individual robot agents. The factors that need to be considered and the architecture used to implement such position selection are also described. Finally the team performance at RoboCup'99 is examined, and future extensions are proposed. 1. Introduction The Robot World Cup initiative (RoboCup) is an attempt to foster AI and intelligent robotics research by providing a uniform task, the game of soccer [4]. Some of the software technologies include design principles of autonomous agents, multi-agent collaboration, strategy acquisitio... | [
1163
] | Validation |
751 | 1 | On the Portability and Tuning of Supervised Word Sense Disambiguation Systems This report describes a set of experiments carried out to explore the portability of alternative supervised Word Sense Disambiguation algorithms. The aim of the work is threefold: firstly, studying the performance of these algorithms when tested on a different corpus from that they were trained on; secondly, exploring their ability to tune to new domains, and thirdly, demonstrating empirically that the LazyBoosting algorithm outperforms state-of-the-art supervised WSD algorithms in both previous situations. Keywords: Word Sense Disambiguation, Machine Learning, Natural Language Processing, Portability and Tuning of NLP systems. 1 Introduction Word Sense Disambiguation (WSD) is the problem of assigning the appropriate meaning (sense) to a given word in a text or discourse where this meaning is distinguishable from other senses potentially attributable to that word. As an example, table 1 shows the definition of two senses of the word age 1 and an example sentence 2 for eac... | [
1066,
1279,
2513,
2676,
2898
] | Train |
752 | 3 | Versus: A Web Repository this paper we consider a Web application (or simply an application), as a Versus client with the ability of executing a task through parallel data processing. Therefore each application should be composed by a group of independent threads | [
1170
] | Train |
753 | 2 | Jotmail: A Voicemail Interface That Enables You to See What Was Said stevew/julia/urs @ research.att.com Voicemail is a pervasive, but under-researched tool for workplace communication. Despite potential advantages of voicemail over email, current phone-based voicemail UIs are highly problematic for users. We present a novel, Web-based, voicemail interface, Jotmail. The design was based on data from several studies of voicemail tasks and user strategies. The GUI has two main elements: (a) personal annotations that serve as a visual analogue to underlying speech; (b) automatically derived message header information. We evaluated Jotmail in an 8-week field trial, where people used it as their only means for accessing voicemail. Jotmail was successful in supporting most key voicemail tasks, although users ' electronic annotation and archiving behaviors were different from our initial predictions. Our results argue for the utility of a combination of annotation based indexing and automatically derived information, as a general technique for accessing speech archives. | [
541
] | Validation |
754 | 3 | Views in a Large Scale XML Repository We are interested in maintaining and querying views in a huge and highly heterogeneous XML repository (Web scale). In this context, views are very large and there is no apparent limitation to their size. This raises interesting problems that we address in the paper: (i) how to distribute views over several machines without having a negative impact on the query translation process; (ii) how to quickly select the relevant part of a view given a query; (iii) how to minimize the cost of communicating potentially large queries to the machines where they will be evaluated. 1 Introduction We believe that XML will soon take an important and increasing share of the data published on the Web. This represents a major opportunity to, at last, provide an intelligent access to this amazing source of information. With that goal in mind, the Xyleme [18] project is building a warehouse which will store and provide sophisticate database-like services over all the XML documents of the Web. Notably... | [
1667,
2621
] | Validation |
755 | 0 | Wasp Nests for Self-Configurable Factories Agent-based approaches to manufacturing scheduling and control have gained increasing attention in recent years. Such approaches are attractive because they offer increased robustness against the unpredictability of factory operations. But the specification of local coordination policies that give rise to efficient global performance and effectively adapt to changing circumstances remains an interesting challenge. In this paper, we introduce a new approach to this coordination problem, drawing on various aspects of a computational model of how wasp colonies coordinate individual activities and allocate tasks to meet the collective needs of the nest. We focus specifically on the problem of configuring machines in a factory to best satisfy (potentially changing) product demands over time. Our system models the set of jobs queued in front of any given machine as a wasp nest, wherein wasp-like agents interact to form a social hierarchy and prioritize the jobs that they represent. Other was... | [
1774
] | Train |
756 | 5 | Algebraic Topology for Knowledge Representation in Analogy Solving . We propose 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 "given A, B and C, find D such that A is to B what C is to D". 1 Introduction In this paper, we present a topological framework for knowledge representation based on the concept of simplicial complex. We present then the ESQIMO system which is the application of this framework to an analogy solving problem. The underlying idea developed here is that spatial relationships and more precisely topological relationships such as neighbor, border, dimension, obstruction, deformation, separabitily, path, etc, enable the building and structuration of knowledge representation. More precisely, we explore the possi... | [
1800
] | Validation |
757 | 5 | Imitation and Mechanisms of Joint Attention: A Developmental Structure for Building Social Skills on a Humanoid Robot Abstract. Adults are extremely adept at recognizing social cues, such as eye direction or pointing gestures, that establish the basis of joint attention. These skills serve as the developmental basis for more complex forms of metaphor and analogy by allowing an infant to ground shared experiences and by assisting in the development of more complex communication skills. In this chapter, we review some of the evidence for the developmental course of these joint attention skills from developmental psychology, from disorders of social development such as autism, and from the evolutionary development of these social skills. We also describe an on-going research program aimed at testing existing models of joint attention development by building a human-like robot which communicates naturally with humans using joint attention. Our group has constructed an upper-torso humanoid robot, called Cog, in part to investigate how to build intelligent robotic systems by following a developmental progression of skills similar to that observed in human development. Just as a child learns social skills and conventions through interactions with its parents, our robot will learn to interact with people using natural social communication. We further consider the critical role that imitation plays in bootstrapping a system from simple visual behaviors to more complex social skills. We will present data from a face and eye finding system that serves as the basis of this developmental chain, and an example of how this system can imitate the head movements of an individual. 1 | [
2197,
2980
] | Train |
758 | 1 | Individual Action and Collective Function: from Sociology to Multi-Agent Learning > How do we characterize the process and the dynamics of co-learning, conceptually, mathematically, or computationally? How do social structures and relations interact with co-learning of multiple agents? 1 And so on. A key question, however, is as follows, which deserves some discussion here. Adam Smith (1976) put it this way: He generally, indeed, neither intends to promote the public interest, nor knows how much he is promoting it..... He intends only his own gain, and he is led by an invisible hand to promote an end which was not part of his intention. This paradox have been troubling sociologists and economists for many decades, and now computer scientists and psychologists as well. The issue may be formulated as the apparent gap between the individual intention in deciding his/her own action and the (possibly largely unintended) social function of his/her action. For example, how may self-interested action benet social welfare? Or, | [
2997
] | Train |
759 | 2 | Learning to Extract Symbolic Knowledge from the World Wide Web The World Wide Web is a vast source of information accessible to computers, but understandable only to humans. The goal of the research described here is to automatically create a computer understandable knowledge base whose content mirrors that of the World Wide Web. Such a knowledge base would enable much more e ective retrieval of Web information, and promote new uses of the Web to support knowledge-based inference and problem solving. Our approach istodevelop a trainable information extraction system that takes two inputs. The rst is an ontology that de nes the classes (e.g., Company, Person, Employee, Product) and relations (e.g., Employed.By, Produced.By) ofinterest when creating the knowledge base. The second is a set of training data consisting of labeled regions of hypertext that represent instances of these classes and relations. Given these inputs, the system learns to extract information from other pages and hyperlinks on the Web. This paper describes our general approach, several machine learning algorithms for this task, and promising initial results with a prototype system that has created a knowledge base describing university people, courses, and research projects. | [
518,
544,
579,
605,
721,
836,
993,
1108,
1226,
1290,
1386,
1532,
1543,
1601,
1676,
1726,
1782,
1823,
1847,
1987,
2054,
2100,
2371,
2510,
2676,
2680,
2817,
2889,
3035
] | Train |
760 | 2 | Using linear classifiers in the integration of user modeling and text content analysis in the personalization of a Web-based Spanish News Service Nowadays many newspapers and news agencies offer personalized information access services and, moreover, there is a growing interest in the improvement of these services. In this paper we present a methodology useful to improve the intelligent personalization of news services and the way it has been applied to a Spanish relevant newspaper: ABC. Our methodology integrates textual content analysis tasks and machine learning techniques to achieve an elaborated user model, which represents separately short-term needs and long-term multi-topic interests. The characterization of a user's interests includes his preferences about structure (newspaper sections), content and information delivery. A wide coverage and non-specific-domain classification of topics and a personal set of keywords allow the user to define his preferences about content. Machine learning techniques are used to obtain an initial representation of each category of the topic classification. Finally, we introduce some details about the Mercurio system, which is being used to implement this methodology for ABC. We describe our experience and an evaluation of the system in comparison with other commercial systems. | [
1857,
2663
] | Train |
761 | 3 | Logical Semantics and Language for Databases with Partial and Complete Tuples and Sets (Extended Abstract) ) Mengchi Liu Department of Computer Science University of Regina, Regina, Saskatchewan, Canada S4S 0A2 Email: mliu@cs.uregina.ca Abstract We discuss the semantics of complex object databases with both partial and complete tuples and sets. We redefine the notion of database to reflect the existence of partial and complete tuples and sets and study how to integrate partial information about tuples and sets spread in the database and check consistency in the meantime. We also present a deductive language RLOG II for complex objects with null/unknown and inconsistent values based on Relationlog. The main novel feature of the language is that it is the only one that supports the null extended nested relational algebra operations directly and more importantly recursively. This work provides a firm logical foundation for nested relational and complex object databases that have both partial and complete tuples and sets and solves an open problem of supporting recursion with generic null/... | [
2083
] | Train |
762 | 4 | The BubbleBadge: A Wearable Public Display We are exploring the design space of wearable computers by designing "public" wearable computer displays. This paper describes our first prototype, the BubbleBadge. By effectively turning the wearer's private display "inside out", the BubbleBadge transforms the wearable computing concept by making digital information public rather than private. User tests showed that the device introduces a new way to interact with information-providing devices, suggesting that it would be valuable to explore the concept further. Keywords Wearable computers, interaction technology, public displays INTRODUCTION A wearable computer is defined as a continuously running, augmenting and mediating computational device [2]. Wearable computers are usually highly private, since both input and output is controlled and seen only by the user, who is effectively "hiding" behind a hand-held keyboard and a head-mounted display. But while wearable computing can be a powerful tool for the single user, there is usuall... | [
2169,
2913
] | Train |
763 | 3 | Data Integration Services Introduction With the prevalence of the network technology and the Internet, access to data independent of its physical storage location has become highly facilitated. This further has enabled users to access a multitude of data sources that are related in some way and to combine the returned data to come up with useful information which is not physically stored in a single place. For instance, a person who has the intension of buying a car can query several car dealer web sites and then compare the results. He can further query a data source which provides information about car reviews to help his decision about the cars he liked. As another example, imagine a company which has several branches in di#erent cities. Each branch has its own local database recording its sales. Whenever global decisions about the company have to be made, each branch database must be queried and the results must be combined. On the other hand, contacting data sources individually and then combining | [
2080
] | Test |
764 | 3 | ROL2: Towards a Real Deductive Object-Oriented Database Language ROL is a strongly typed deductive object-oriented database language. It integrates many important features of deductive databases and object-oriented databases. However, it is only a structurally object-oriented language. In this paper, we present our extension of ROL called ROL2. Most importantly, ROL2 supports behaviorally objectoriented features such as rule-based methods and encapsulation so that it is a now real deductive object-oriented database language. It supports in a rule-based framework nearly all important object-oriented features such as object identity, complex objects, typing, information hiding, rule-based methods, encapsulation of such methods, overloading, late binding, polymorphism, class hierarchies, multiple structural and behavioral inheritance with overriding, blocking, and conict handling. It is so far the only deductive system that supports all these features in a pure rule-based framework. Keywords: object-oriented databases, deductive databases, ... | [
178,
2623
] | Train |
765 | 4 | Virtual Keyboards This paper describes a novel scheme for vision-based human computer interaction in which traditional input and output devices, monitors, keyboards and mice, are replaced with augmented reality displays, projection systems and cameras. User input is accomplished by projecting an image of the interface onto a flat surface in the scene which is monitored with a video camera. The scheme hinges on the observation that the relationship between the three surfaces of interest, the work surface, the virtual keyboard and the image obtained by the camera, can be characterized by projective transformations of RP 2 . This observation leads to a fast and accurate online calibration algorithm. The basic advantage of the vision based interaction techniques proposed in this paper is that they do not involve mechanical input devices such as keyboards, mice and touch screens. There are no moving parts and no wires to connect to the interface surface. By avoiding a physical instantiation of t... | [
727,
3155
] | Validation |
766 | 4 | Adaptable and Adaptive Information Provision for All Users, Including Disabled and Elderly People Due to the tremendously increasing popularity of the World-Wide Web, hypermedia is going to be the leading online information medium for some years to come and will most likely become the standard gateway for citizens to the "information highway". Already today, visitors of web sites are generally heterogeneous and have different needs, and this is likely to increase in the future. The aim of the AVANTI project is to cater hypermedia information to these individual needs by adapting the content and the presentation of web pages to each individual user. The special needs of elderly and disabled users are also partly considered. A model of the characteristics of user groups, individual users and usage environments, and a domain model are exploited in the adaptation process. One aim of this research is to verify that adaptation and user modeling techniques that were hitherto mostly used for catering interactive software systems to able-bodied users also prove useful for adaptation to users with special needs. Another original aspect is the development of a network-wide user modeling server that can concurrently accommodate the user modeling needs of several applications and several instances of an application within a distributed computing environment. | [
1892
] | Train |
767 | 0 | Autonomous Robot that Uses Symbol Recognition and Artificial Emotion to Attend the AAAI Conference This paper describes our approach in designing an autonomous robot for the AAAI Mobile Robot Challenge, making the robot attend the National Conference on AI. The goal was to do a simplified version of the whole task, by integrating methodologies developed in various research projects conducted in our laboratory. Original contributions are the use of a symbol recognition technique to make the robot read signs, artificial emotion for expressing the state of the robot in the accomplishment of its goals, a touch screen for human-robot interaction, and a charging station for allowing the robot to recharge when necessary. All of these aspects are influenced by the different steps to be followed by the robot attendee to complete the task from start-to-end. Introduction LABORIUS is a young research laboratory interested in designing autonomous systems that can assist human in real life tasks. To do so, robots need some sort of "social intelligence ", giving them the ability to ... | [
963,
2828
] | Train |
768 | 3 | Efficient Concurrency Control for Broadcast Environments A crucial consideration in environments where data is broadcast to clients is the low bandwidth available for clients to communicate with servers. Advanced applications in such environments do need to read data that is mutually consistent aswell as current. However, given the asymmetric communication capabilities and the needs of clients in mobile environments, traditional serializability-based approaches are too restrictive, unnecessary, and impractical. We thus propose the use of a weaker correctness criterion called update consistency and outline mechanisms based on this criterion that ensure (1) the mutual consistency of data maintained by the server and read by clients, and (2) the currency of data read by clients. Using these mechanisms, clients can obtain data that is current and mutually consistent "off the air", i.e., without contacting the server to, say, obtain locks. Experimental results show a substantial reduction in response times as compared to existing (serializability-based) approaches. A further attractive feature of the approach is that if caching is possible at a client, weaker forms of currency can be obtained while still satisfying the mutual consistency of data. | [
381,
800,
954,
1047,
1131
] | Train |
769 | 5 | Ensemble Learning for Intrusion Detection in Computer Networks The security of computer networks plays a strategic role in modern computer systems. In order to enforce high protection levels against threats, a number of software tools are currently developed. Intrusion Detection Systems aim at detecting intruder who eluded the "first line" protection. In this paper, a pattern recognition approach to network intrusion detection based on ensemble learning paradigms is proposed. The potentialities of such an approach for data fusion and some open issues are outlined. | [] | Train |
770 | 0 | Planning Agents in James Abstract — Testing is an obligatory step in developing multi-agent systems. For testing multi-agent systems in virtual, dynamic environments, simulation systems are required that support a modular, declarative construction of experimental frames, that facilitate the embeddence of a variety of agent architectures, and that allow an efficient parallel, distributed execution. We introduce the system James (A Java-Based Agent Modeling Environment for Simulation). In James agents and their dynamic environment are modeled as reflective, time triggered state automata. Its possibilities to compose experimental frames based on predefined components, to express temporal interdependencies, to capture the phenomenon of pro-activeness and reflectivity of agents are illuminated by experiments with planning agents. The underlying planning system is a general purpose system, about which no empirical results exist besides traditional static benchmark tests. We analyze the interplay between heuristics for selecting goals, viewing range, commitment strategies, explorativeness, and trust in the persistence of the world and uncover properties of the agent, the planning engine and the chosen test scenario: Tileworld. I. | [
12,
2343,
2357,
2402,
2703,
3173
] | Train |
771 | 5 | A Multi-Agent Architecture for Intelligent Tutoring One of the most interesting realm among those ones brought up to success by the development of the Internet is Distance Learning. A key issue in such a field is the development of systems for supporting Tutoring activities. This paper is concerned with the presentation of an innovative architecture for Intelligent Tutoring which make use of Software Agents. The way in which the knowledge is represented and stored is discussed together with the ability of our system to manage individual learning paths for different users. The rationale for using Agents is presented and the implementation of the system is discussed. 1. | [
1001,
2229
] | Train |
772 | 0 | An Architecture to Guide Crowds Using a Rule-Based Behavior System This paper describes a Client/Server architecture to combine the control of human agents performing "intelligent actions" (guided by a Rule-Based Behavior System -- RBBS) with the management of autonomous crowds which perform pre-programmed actions. Our main goal being ability to model crowds formed by a large number of agents (e.g. 1000), we have used pre-programmed actions and basic behaviors. In addition, RBBS provides the user with an interface for real-time behavior control of some groups of the crowd. This paper presents how the Server application deals with virtual human agent's behaviors using a rule-based system. Keywords Multi-agent co-ordination and collaboration, agent architectures, network agents, real-time performance, synthetic agents, rulebased system, human crowds' model. 1. INTRODUCTION Virtual humans grouped together to form crowds populating virtual worlds allow a more intuitive feeling of presence. However, the crowd is not only needed to create an at... | [
1548
] | Test |
773 | 1 | Parameter-less Genetic Algorithm: A Worst-case Time and Space Complexity Analysis In this paper, the worst-case analysis of the time and space complexity of the parameter-less genetic algorithm versus the genetic algorithm with an optimal population size is provided and the results of the analysis are discussed. Since the assumptions in order for the analysis to be correct are very weak, the result is applicable to a wide range of problems. Various configurations of the parameter-less genetic algorithm are considered and the results of their time and space complexity are compared. 1 Introduction A parameter-less genetic algorithm (Harik & Lobo, 1999) is an alternative to a common trialand -error method of tweaking the values of the parameters of the genetic algorithm in order to find a set-up to accurately and reliably solve a given problem. The algorithm manages a number of independent runs of the genetic algorithm with different population sizes with the remaining parameters set to fixed values according to the theory of genetic algorithms' control maps introduce... | [
2579
] | Train |
774 | 0 | The Repair of Speech Act Misunderstandings by Abductive Inference this paper, we have concentrated on the repair of mis-understanding. Our colleagues Heeman and Edmonds have looked at the repair of non-understanding. The difference between the two situations is that in the former, the agent derives exactly one interpretation of an utterance and hence is initially unaware of any problem; in the latter, the agent derives either more than one interpretation, with no way to choose between them, or no interpretation at all, and so the problem is immediately apparent. Heeman and Edmonds looked in particular at cases in which a referring expression uttered by one conversant was not understood by the other (Heeman and Hirst 1995; Edmonds 1994; Hirst et. al. 1994). Clark and his colleagues (Clark and Wilkes-Gibbs 1986; Clark 1993) have shown that in such situations, conversants will collaborate on repairing the problem by, in effect, negotiating a reconstruction or elaboration of the referring expression. Heeman and Edmonds model this with a plan recognition and generation system that can recognize faulty plans and try to repair them. Thus (as in our own model) two copies of the system can converse with each other, negotiating referents of referring expressions that are not understood by trying to recognize the referring plans of the other, repairing them where necessary, and presenting the new referring plan to the other for approval. | [
395,
527,
1008,
1629,
3022
] | Test |
775 | 3 | SENTINEL: A Multiple Engine Information Retrieval and Visualization System We describe a prototype Information Retrieval system, SENTINEL, under development at Harris Corporation's Information Systems Division. SENTINEL is a fusion of multiple information retrieval technologies, integrating n-grams, a vector space model, and a neural network training rule. One of the primary advantages of SENTINEL is its 3-dimenstional visualization capability that is based fully upon the mathematical representation of information within SENTINEL. This 3-dimensional visualization capability provides users with an intuitive understanding, with relevance feedback/query refinement techniques that can be better utilized, resulting in higher retrieval accuracy (precision). | [
1990
] | Validation |
776 | 0 | Foreign Event Handlers to Maintain Information Consistency and System Adequacy this paper is to describe novel applications of Mobile Code technology which have not appeared yet but should be feasible with our current knowledge of the domain. These new applications contradict the often-made observation that Mobile Code is just another technique that does not really bring much more possibilities than existing technologies for distributed applications. There is a whole class of problems that have not received much attention yet and that are not well managed by current environments. These are the problems of maintaining consistency of dynamic information and maintaining systems in adequacy with the ever changing requirements of customers. Our motivation is that, besides the quantitative improvements that most people expect from using Mobile Code, there is also a qualitative benefit which is even more important but not universally recognized now: Mobile Code allows communication with less conventions than message passing [5, 3]. Processes interconnected by Mobile Code still have to agree on high level encoding and synchronization primitives but these agreements are only a fraction of what is necessary to communicate. Many context dependent aspects can be encapsulated inside Mobile Code and changed when the context changes. Encapsulation has the same benefits here as in other software engineering domains: it reduces the dependency between components, thus reducing the number of modifications that we must make to software in order to adapt it to new requirements. For this reason we think that it is the best way to cope with systems that are distributed, hence not manageable by a single person or organization; that are dynamic, because the information they contain must change when the world itself changes; and that are evolving since the users discover n... | [
2619
] | Validation |
777 | 0 | Introducing Trusted Third Parties to the Mobile Agent Paradigm Abstract. The mobile agent paradigm gains ever more acceptance for the creation of distributed applications, particularly in the domain of electronic commerce. In such applications, a mobile agent roams the global Internet in search of services for its owner. One of the problems with this approach is that malicious service providers on the agent's itinerary can access con dential information contained in the agent or tamper with the agent. In this article we identify trust as a major issue in this context and propose a pessimistic approach to trust that tries to prevent malicious behaviour rather than correcting it. The approach relies on a trusted and tamper-resistant hardware device that provides the mobile agent with the means to protect itself. Finally, weshow that the approach is not limited to protecting the mobile agents of a user but can also be extended to protect the mobile agents of a trusted third party inorder to take fulladvantage of the mobile agent paradigm. 1 | [
169
] | Train |
778 | 5 | A Competitive Layer Model for Feature Binding and Sensory Segmentation We present a recurrent neural network for feature binding and sensory segmentation, the competitive layer model (CLM). The CLM uses topographically structured competitive and cooperative interactions in a layered network to partition a set of input features into salient groups. The dynamics is formulated within a standard additive recurrent network with linear threshold neurons. Contextual relations among features are coded by pairwise compatibilities which define an energy function to be minimized by the neural dynamics. Due to the usage of dynamical winner-take-all circuits the model gains more flexible response properties than spin models of segmentation by exploiting amplitude information in the grouping process. We prove analytic results on the convergence and stable attractors of the CLM, which generalize earlier results on winner-take-all networks, and incorporate deterministic annealing for robustness against local minima. The piecewise linear dynamics of the CLM allows a linear eigensubspace analysis which we use to analyze the dynamics of binding in conjunction with annealing. For the example of contour detection we show how the CLM can integrate figure-ground segmentation and grouping into a unified model. | [
2695
] | Train |
779 | 0 | An Agent Infrastructure to Build and Evaluate Multi-Agent Systems: The Java Agent Framework and Multi-Agent System Simulator . In this paper, we describe our agent framework and address | [
1107,
2043,
2358
] | Test |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.