node_id int64 0 76.9k | label int64 0 39 | text stringlengths 13 124k | neighbors listlengths 0 3.32k | mask stringclasses 4
values |
|---|---|---|---|---|
1,280 | 2 | Learning from Labeled and Unlabeled Data using Graph Mincuts Many application domains suffer from not having enough labeled training data for learning. However, large amounts of unlabeled examples can often be gathered cheaply. As a result, there has been a great deal of work in recent years on how unlabeled data can be used to aid classification. We consider an algorithm based on finding minimum cuts in graphs, that uses pairwise relationships among the examples in order to learn from both labeled and unlabeled data. | [
855,
3078
] | Train |
1,281 | 3 | Widening Pos for Efficient and Scalable Groundness Analysis of Logic Programs The domain of positive Boolean functions, Pos, is by now well established for the analysis of the variable dependencies that arise within logic programs. Analyses based on Pos that use binary decision diagrams have been shown to be efficient for a wide range of practical programs. However, independent of the representation, assuming that P != NP , an (unwidened) Pos analysis can never come with any efficiency guarantees because of its potentially explosive behaviour. This paper proposes a simple widening for (goal-dependent) groundness analysis of logic programs that guarantees scalability and efficiency. Experimental results indicate that the widening induces only a very small loss of precision. | [
2413
] | Validation |
1,282 | 0 | Formalisms for Multi-Agent Systems This report is the result of a panel discussion at the First UK Workshop on Foundations of Multi-Agent Systems (FoMAS '96). All members of the panel are authors, listed alphabetically. as knowledge representation language, for direct manipulation within an agent system, is exemplified in the work of Konolige on formalisms for modelling belief, and logic as a programming language is evidenced in the work of Fisher on Concurrent METATEM. All of these strands of work can claim some measure of success. However, a common failing of formal work (both in AI and multi-agent systems) is that its role is not clear. Formal agent theories are agent specifications, not only in the sense of providing descriptions and constraints on agent behaviour, but also in the sense that one understands the term `specification' from mainstream software engineering, namely that they provide a base from which to design, implement and verify agent systems. Agents are a natural next step for software engineering; they represent a fundamentally new way of considering complex distributed systems, containing societies of cooperating autonomous components. If we aim to build such systems, then principled techniques will be required for their design and implementation. We aim to assist the development of such systems by providing formalisms and notations that can be used to specify the desirable behaviour of agents and multi-agent systems; a requirement is that we should be able to move in a principled way from specifications of such systems to implementations. The properties identified by using a formalism serve to measure and evaluate implementations of agent systems. Some properties currently seem to be unimplementable, because they deal with an idealised aspect of agency, such as knowledge. Still, t... | [
1297
] | Validation |
1,283 | 3 | Data Mining for Intelligent Web Caching The paper presents a vertical application of data warehousing and data mining technology: intelligent web caching. We introduce several ways to construct intelligent web caching algorithms that employ predictive models of web requests; the general idea is to extend the LRU policy of web and proxy servers by making it sensible to web access models extracted from web log data using data mining techniques. Two approaches have been studied in particular, one based on association rules and another based on decision trees. The experimental results of the new algorithms show substantial improvement over existing LRUbased caching techniques, in terms of hit rate, i.e., the fraction of web documents directly retrieved in the cache. We designed and developed a prototypical system, which supports data warehousing of web log data, extraction of data mining models and simulation of the web caching algorithms, around an architecture that integrates the various phases in the knowledge discovery process. The system supports a systematic evaluation and benchmarking of the proposed algorithms with respect to existing caching strategies. 1 | [
255
] | Train |
1,284 | 1 | On Concept Space and Hypothesis Space in Case-Based Learning Algorithms . In order to learn more about the behaviour of case-based reasoners as learning systems, we formalise a simple case-based learner as a PAC learning algorithm. We show that the case-based representation hCB; oei is rich enough to express any boolean function. We define a family of simple case-based learning algorithms which use a single, fixed similarity measure and we give necessary and sufficient conditions for the consistency of these learning algorithms in terms of the chosen similarity measure. Finally, we consider the way in which these simple algorithms, when trained on target concepts from a restricted concept space, often output hypotheses which are outside the chosen concept space. A case study investigates this relationship between concept space and hypothesis space and concludes that the case-based algorithm studied is a less than optimal learning algorithm for the chosen, small, concept space. 1 Introduction The performance of a case-based reasoning system [13] will chan... | [
419,
1424,
2581,
2878,
3160
] | Test |
1,285 | 0 | Intrusion Detection: A Bibliography This document contains more than 600 references, dated from 1980 to 2001. We undoubtedly have forgotten some important citations, either through oversight or ignorance. Moreover, errors may remain in the citations. Thus, we ask for your indulgence and, more importantly, for your help. Send us a note if you nd any errors and let us know of any omissions | [
123,
1240,
1335,
1411,
2052,
2833
] | Validation |
1,286 | 1 | Learning To Locate An Object in 3D Space From A Sequence Of Camera Images This paper addresses the problem of determining an object's 3D location from a sequence of camera images recorded by a mobile robot. The approach presented here allows people to "train" robots to recognize specific objects, by presenting it examples of the object to be recognized. A decision tree method is used to learn significant features of the target object from individual camera images. Individual estimates are integrated over time using Bayes rule, into a probabilistic 3D model of the robot's environment. Experimental results illustrate that the method enables a mobile robot to robustly estimate the 3D location of objects from multiple camera images. 1 INTRODUCTION In recent years, there has been significant progress in the field of mobile robotics. Applications such as robots that guide blind or mentally handicapped people, robots that clean large office buildings and department stores, robots that assist people in recreational activities, etc., are slowly getting in reach. Man... | [
2547,
2999
] | Train |
1,287 | 4 | Ubiquitous Web Information Agents . This paper gives a brief overview about the AI methods and techniques we have developed for building ubiquitous web information systems. These methods from areas of machine learning, logic programming, knowledge representation and multi-agent systems are discussed in the context of our prototypical information system MIA. MIA is a web information system for mobile users, who are equipped with a PDA (Palm Pilot), a cellular phone and a GPS device or cellular WAP phone. It captures the main issues of ubiquitous computing: location awareness, anytime information access and PDA technology. 1 Introduction Nowadays, the biggest but also the most chaotic and unstructured source of information is the World-Wide-Web. Making this immense amount of information available for ubiquitous computing in daily life is a great challenge. Besides hardware issues for wireless ubiquitous computing, that still are to be solved (wireless communication, blue-tooth technologies, wearable computing u... | [
327,
439,
1569
] | Validation |
1,288 | 1 | Cooperative Coevolution of Multi-Agent Systems In certain tasks such as pursuit and evasion, multiple agents need to coordinate their behavior to achieve a common goal. An interesting question is, how can such behavior best be evolved? When the agents are controlled with neural networks, a powerful method is to coevolve them in separate subpopulations, and test together in the common task. In this paper, such a method, called Multi-Agent ESP (Enforced Subpopulations) is presented, and demonstrated in a prey-capture task. The approach is shown more efficient and robust than evolving a single central controller for all agents. The role of communication in such domains is also studied, and shown to be unnecessary and even detrimental if effective behavior in the task can be expressed as role-based cooperation rather than synchronization. 1 | [
1522
] | Validation |
1,289 | 0 | Analysis and Design using MaSE and agentTool This paper provides an overview of the work being done at the Air Force Institute of Technology on the Multiagent Systems Engineering methodology and the associated agentTool environment. Our research is focused on discovering methods and techniques for engineering practical multiagent systems. It uses the abstraction provided by multiagent systems for developing intelligent, distributed software systems. | [
1172,
1297,
1681
] | Train |
1,290 | 1 | Combining Labeled and Unlabeled Data with Co-Training We consider the problem of using a large unlabeled sample to boost performance of a learning algorithm when only a small set of labeled examples is available. In particular, we consider a setting in which the description of each example can be partitioned into two distinct views, motivated by the task of learning to classify web pages. For example, the description of a web page can be partitioned into the words occurring on that page, and the words occurring in hyperlinks that point to that page. We assume that either view of the example would be su cient for learning if we had enough labeled data, but our goal is to use both views together to allow inexpensive unlabeled data to augment amuch smaller set of labeled examples. Speci cally, the presence of two distinct views of each example suggests strategies in which two learning algorithms are trained separately on each view, and then each algorithm's predictions on new unlabeled examples are used to enlarge the training set of the other. Our goal in this paper is to provide a PAC-style analysis for this setting, and, more broadly, a PAC-style framework for the general problem of learning from both labeled and unlabeled data. We also provide empirical results on real web-page data indicating that this use of unlabeled examples can lead to signi cant improvement of hypotheses in practice. As part of our analysis, we provide new re- | [
526,
759,
855,
888,
1478,
2176,
2178,
2610,
2682,
2808,
2918,
3035
] | Train |
1,291 | 0 | Leveled Commitment Contracts and Strategic Breach In automated negotiation systems for self-interested agents, contracts have traditionallybeen binding. Theydo not accommodate future events. Contingency contracts address this, but are often impractical. As an alternative, we propose le�eled commitment contracts. The level of commitment is set bybreach penalties. To be freed from the contract, an agent simply pays the penalty to the other party. A self-interested agent will be reluctant to breach because the other partymight breach, in which case the former agent is freed from the contract, does not incur a penalty, and collects a penalty from the breacher. We show that, despite such strategic breach, leveled commitment increases the expected payoff to both contract parties and can enable deals that are impossible under full commitment. Asymmetric beliefs are also discussed. Different decommitting mechanisms are introduced and compared. Practical prescriptions for market designers are provided. A contract optimizer is provided on the web. Journal ofEconomic Literature | [
1658,
2465
] | Train |
1,292 | 5 | Embodied Cassie We have enhanced a computational cognitive agent by embodying it with real and simulated bodies operating in real and simulated worlds. This has allowed us to experiment with various ways that embodiment influences the creation and meaning of the agent's beliefs and other terms in its knowledge base, including: symbol-grounding by perception and action; firstperson privileged knowledge; the representation and use of indexicals; having a personal sense of time; and low-level bodily awareness. Introduction We have been engaged in a series of projects in which Cassie, the SNePS cognitive agent (Shapiro & Rapaport 1987; Shapiro 1989; Shapiro & Rapaport 1991; 1992; Shapiro & The SNePS Implementation Group 1998), has been incorporated into a hardware or softwaresimulated cognitive robot. The capabilities of the embodied Cassie have included: input and output in fragments of English; reasoning; performance of primitive and composite acts; and vision. In this paper, I give an overview of the... | [
715
] | Validation |
1,293 | 0 | ATNoSFERES: a Model for Evolutive Agent Behaviors This paper introduces ATNoSFEERS, a model aimed at designing evolutive and adaptive behaviors for agents or multi-agent systems. We first discuss briefly the main problems raised by classical evolutionary models, which are not intended to produce agents or behaviors but to solve problems. Then we provide detailed explanations about the model we propose and its components. We also show through a simple example how the system works, and give some experimental results. Finally, we discuss the features of our model and propose extensions. | [
1473
] | Validation |
1,294 | 2 | Finite-state approaches to Web information extraction Introduction An information agent is a distributed system that receives a goal through its user interface, gathers information relevant to this goal from a variety of sources, processes this content as appropriate, and delivers the results to the users. We focus on the second stage in this generic architecture. We survey a variety of information extraction techniques that enable information agents to automatically gather information from heterogeneous sources. For example, consider an agent that mediates package-delivery requests. To satisfy such requests, the agent might need to retrieve address information from geographic services, ask an advertising service for freight forwarders that serve the destination, request quotes from the relevant freight forwarders, retrieve duties and legal constraints from government sites, get weather information to estimate transportation delays, etc. Information extraction (IE) is a form of shallow document processing that involves populating | [
603,
906,
2068,
2418,
2466,
3053,
3099
] | Test |
1,295 | 2 | Developing Language Processing Components with GATE (a User Guide) Contents 1 Introduction 3 1.1 How to Use This Text . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2 Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.4 Structure of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 1.5 Further Reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 2 How To. . . 14 2.1 Download GATE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.2 Install and Run GATE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.3 [D,F] Configure GATE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 2.4 Build GATE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.5 [D,F] Create a New CREOLE Resource . . . . . . . . . . . . . . . . . . . . 18 2.6<F11 | [
703,
1363,
2123,
2663
] | Train |
1,296 | 2 | Visual Ranking of Link Structures (Extended Abstract) Methods for ranking World Wide Web resources according to their position in the link structure of the Web are receiving considerable attention, because they provide the first effective means for search engines to cope with the explosive growth and diversification of the Web. | [
568,
1838,
1959,
2459,
2503
] | Train |
1,297 | 0 | A Survey of Agent-Oriented Methodologies . This article introduces the current agent-oriented methodologies. It discusseswhat approacheshave been followed (mainly extending existing objectoriented and knowledge engineering methodologies), the suitability of these approaches for agent modelling, and some conclusions drawn from the survey. 1 Introduction Agent technology has received a great deal of attention in the last few years and, as a result, the industry is beginning to get interested in using this technology to develop its own products. In spite of the different developed agent theories, languages, architectures and the successful agent-based applications, very little work for specifying (and applying) techniques to develop applications using agent technology has been done. The role of agent-oriented methodologies is to assist in all the phases of the life cycle of an agent-based application, including its management. This article reviews the current approaches to the development of an agent-oriented (AO) methodology. ... | [
245,
320,
1012,
1157,
1282,
1289,
1414,
1759,
2164,
2659,
2747,
3000,
3132
] | Train |
1,298 | 2 | Keys for XML this paper. The proposal extends the key speci cation of XML Data by allowing one to specify keys in terms of XPath [24] expressions. There are a number of technical problems in connection with XPath. XPath is a relatively complex language in which one can not only move down the document tree, but also sideways or upwards, not to mention that predicates and functions can be embedded as well. The problem with XPath is that questions about equivalence or inclusion of XPath expressions are, as far as the authors are aware, unresolved; and these issues are importantifwewant to reason about keys as wedo in relational databases. Yet until we know how to determine the equivalence of XPath expressions, there is no general method of saying whether two such speci cations are equivalent. Another technical issue is value equality. XML Schema restricts equality to text, but the authors have encountered cases in whichkeys are not so restricted. See Section 7.1 for a more detailed discussion. However, the main reason for writing this paper is that none of the existing key proposals address the issue of hierarchical keys, which appear to be ubiquitous in hierarchically structured databases, especially in scienti c data formats. A top-level key may be used to identify components of a document, and within each component a secondary key is used to identify sub-components, and so on. Moreover, the authors believe that the use of keys for citing parts of a document is suciently important that it is appropriate to consider key speci cation independently of other proposals for constraining the structure of XML documents | [
1318
] | Test |
1,299 | 1 | An asymptotic analysis of AdaBoost in the binary classification case Recent work has shown that combining multiple versions of weak classifiers such as decision trees or neural networks results in reduced test set error. To study this in greater detail, we analyze the asymptotic behavior of AdaBoost type algorithms. The theoretical analysis establishes the relation between the distribution of margins of the training examples and the generated voting classification rule. The paper shows asymptotic experimental results for the binary classification case underlining the theoretical findings. Finally, the relation between the model complexity and noise in the training data, and how to improve AdaBoost type algorithms in practice are discussed. 1 Introduction An ensemble is a collection of neural networks or other types of classifiers (predictors) that are trained for the same task. Boosting and other ensemble learning methods have been used recently with great success for several applications, e. g. OCR [6, 4]. In this work we investigate the functioning o... | [
723,
2238
] | Validation |
1,300 | 1 | Hierarchical Models for Screening of Iron Deficiency Anemia We investigate the problem of classifying individuals based on estimated density functions for each individual. The problem is similar to conventional classification in that there is labelled training data, but different in that the underlying measurements are not feature vectors but histograms or density estimates. We describe a general framework based on probabilistic hierarchical models for modelling such data and illustrate how the model lends itself to classification. We contrast this approach with two other alternatives: (1) directly defining distance between densities using a cross-entropy distance measure, and (2) using parameters of the estimated densities as feature vectors for a standard discriminative classification framework. We evaluate all three methods on a realworld medical diagnosis problem. The hierarchical modeling and density-distance approaches are most accurate, yielding crossvalidated error rates in the range of 1 to 2%. We conclude by discussing the relative me... | [
3007
] | Train |
1,301 | 1 | Constructing Qualitative Event Models Automatically from Video Input We describe an implemented technique for generating event models automatically based on qualitative reasoning and a statistical analysis of video input. Using an existing tracking program which generates labelled contours for objects in every frame, the view from a fixed camera is partitioned into semantically relevant regions based on the paths followed by moving objects. The paths are indexed with temporal information so objects moving along the same path at different speeds can be distinguished. Using a notion of proximity based on the speed of the moving objects and qualitative spatial reasoning techniques, event models describing the behaviour of pairs of objects can be built, again using statistical methods. The system has been tested on a traffic domain and learns various event models expressed in the qualitative calculus which represent human observable events. The system can then be used to recognise subsequent selected event occurrences or unusual behaviours. 1 Introduction D... | [
143,
2420
] | Train |
1,302 | 4 | Topographic Maps Based on Kohonen Self Organizing Maps An Empirical Approach Two-dimensional maps are a valuable interface element for the visualization of information retrieval results or other large sets of objects. Various methods exist for the creation of these maps. This article describes a comparative evaluation of topographic maps based on "Kohonen Self Organizing Maps" (SOM). These results show that the mapping method has to be chosen very carefully and different methods should be tested for an application. KEYWORDS: self organizing maps, Kohonen maps, topic maps, 2D maps, evaluation, information retrieval, information visualization | [
87
] | Train |
1,303 | 3 | Planned Disconnections for Mobile Databases As mobility permeates todays computing environment, we envision application infrastructures that will increasingly use mobile technologies. Traditional database applications will need to integrate mobile entities: people and computers. In this paper, we develop a distributed database framework for mobile environments. A key requirement in such an environment is to support frequent connection and disconnection of database sites. 1 Introduction As mobility permeates into todays computing and communication arena, we envision application infrastructures that will increasingly rely on mobile technologies. Current mobility applications tend to have a large central server and use mobile platforms only as caching devices. We want to elevate the role of mobile computers to first class entities in the sense that they allow the mobile user work/update capabilities independent of a central server. In such an environment, several mobile computers may collectively form the entire distributed syste... | [
2183
] | Validation |
1,304 | 4 | Roomware: Towards the next generation of human-computer interaction based on an integrated design of real and virtual worlds In the past, a central mainframe computer provided terminals for many users. In the current age of the personal desktop computer, there is one computer for one person. Observation of early adopters and predictions about the future point to an era where each person will have multiple devices and computational power will be ubiquitous. Against this background, we present a vision for the workspaces of the future and a user-centered approach for an integrated design of virtual information spaces and real architectural spaces. The resulting environments are called cooperative buildings. The design approach is based on the roomware concept. By roomware, we mean computer-augmented objects resulting from the integration of room elements, e.g., walls, doors, furniture (tables, chairs, etc.) with computer-based information devices. They are part of the vision that the world around us will be the interface to information -- where the computer as a device will disappear and people's interaction w... | [
682,
708,
2511
] | Test |
1,305 | 0 | Secure Dynamic Reconfiguration of Scalable CORBA Systems with Mobile Agents . As various Internet services, electronic commerce, and information and communication systems permeate our lives, their continual availability becomes a dominant issue. But continuing software evolution requires system recon guration. Running systems must upgrade their components or change their con guration parameters. In addition, Internet services often need to serve thousands or millions of users. This scenario raises three conicting issues that need to be resolved: availability, con gurability, and scalability. We propose the use of mobile recon guration agents for the ecient, secure, and scalable dynamic recon guration of Internet systems. In our system, the agents are deployed within the object-oriented architecture of a CORBA-compliant ORB that supports safe recon guration both of its middleware engine and of user applications. Using a graphical front-end, administrators build recon guration agents and specify the topology of the recon guration network the agents traverse to recon gure the distributed system on-the-y. The agents collect, group, and return the results of inspection and recon guration operations in the distributed system to the administrator. 1 | [
1739
] | Train |
1,306 | 5 | HYMES: A HYbrid Modular Expert System with Efficient Inference and Explanation A HYbrid Modular Expert System, called HYMES, is presented. HYMES provides a dual representation scheme: a symbolic one, based on conventional symbolic rules, and a hybrid one, based on neumles, a kind of rules that combine a symbolic and a connectionist representation. Symbolic rules are internally converted into neumles, for efficiency reasons. In this way, hybrid modular knowledge bases can be constructed. | [
3165
] | Train |
1,307 | 2 | An Approach to Relate the Web Communities Through Bipartite Graphs The Web harbors a large number of community structures. Early detection of community structures has many purposes such as reliable searching and selective advertising. In this paper we investigate the problem of extracting and relating the web community structures from a large collection of Web-pages by performing hyper-link analysis. The proposed algorithm extracts the potential community signatures by extracting the corresponding dense bipartite graph (DBG) structures from the given data set of web pages. Further, the proposed algorithm can also be used to relate the extracted community signatures. We report the experimental results conducted on 10 GB TREC (Text REtrieval Conference) data collection that contains 1.7 million pages and 21.5 million links. The results demonstrate that the proposed approach extracts meaningful community signatures and relates them. | [
1201,
1838,
1983,
2459,
2503,
3077
] | Train |
1,308 | 2 | Discovery of Similarity Computations of Search Engines Two typical situations in which it is of practical interest to determine the similarities of text documents to a query due to a search engine are: (1) a global search engine, constructed on top of a group of local search engines, wishes to retrieve the set of local documents globally most similar to a given query; and (2) an organization wants to compare the retrieval performance of search engines. The dot-product function is a widely used similarity function. For a search engine using such a function, we can determine its similarity computations if how the search engine sets the weights of terms is known, which is usually not the case. In this paper, techniques are presented to discover certain mathematical expressions of these formulas and the values of embedded constants when the dot-product similarity function is used. Preliminary results from experiments on the WebCrawler search engine are given to illustrate our techniques. 1 Categories and Subject Descriptors H.3 [Information... | [
521,
1134,
1167,
1642
] | Train |
1,309 | 0 | The EASE Actor Development Environment In interactive simulations it is often desirable to have intelligent actors playing the roles of humans. Drawing on a wide range of previous work this paper presents a system that is intended to reduce some of the diculties involved in the development of actors. We present a system called EASE (End-user Actor Specication Environment) that provides tools and methods to support end user development of intelligent actors. The tools support the whole development process from design to testing. The EASE actor architecture is a multi-agent system where a process of contract making and negotiation between agents determines the actions of the actor. 1 Introduction In modern, complex, interactive simulations it is often highly desirable to have intelligent actors playing the roles of humans. The actors' task is dicult { sensing the (simulated) environment, choosing a course of action that exibly and intelligently follows designer intentions and sending appropriate commands back to ... | [
1151,
1943,
2309
] | Train |
1,310 | 0 | Modeling Emotion-Based Decision-Making This paper presents a computational approach to EmotionBased Decision-Making that models important aspects of emotional processing and integrates these with other models of perception, motivation, behavior, and motor control. A particular emphasis is placed on using some of the mechanisms of emotions as building blocks for the acquisition of emotional memories that serve as biasing signals during the process of making decisions and selecting actions. We have successfully followed this approach to develop and control several different autonomous agents, including both synthetic agents and physical robots. Introduction Most theories of human reasoning and decision-making fall between two different positions. The first one argues that we make decisions in a way similar to that of solving problems in formal logic. According to this view, when faced with a problem, we form a list of all different options and their possible outcomes, and then we use logic in its best sense to perform a cos... | [
115
] | Test |
1,311 | 3 | Definition And Analysis Of Index Organizations For Object-Oriented Database Systems The efficient execution of queries in object-oriented databases requires the design of specific indexing techniques, to efficiently deal with predicates against nested attributes or against class inheritance hierarchies. Indexing techniques so far proposed can be classified into three groups: inheritance indexing techniques, whose goal is to support queries along inheritance hierarchies; aggregation indexing techniques, dealing with the efficient evaluation of nested predicates; integrated techniques. The aim of this paper is to analyze two techniques providing an integrated support, the path index and the nested-inherited index, with respect to traditional techniques, such as the multi-index and the inherited multi-index. The analysis is performed assuming that multi-valued attributes, as well as instances with null attribute values, are present in the database. For this purpose, the paper first presents the considered techniques. An extension of the path index, firstly defined in [6]... | [] | Train |
1,312 | 2 | Incremental Document Clustering for Web Page Classification Motivated by the benefits in organizing the documents in Web search engines, we consider the problem of automatic Web page classification. We employ the clustering techniques. Each document is represented by a feature vector. By analyzing the clusters formed by these vectors, we can assign the documents within the same cluster to the same class automatically. Our contributions are the following: (1) We propose a feature extraction mechanism which is more suitable to Web page classification. (2) We introduce a tree structure called the DC-tree to make the clustering process incremental and less sensitive to the document insertion order. (3) We show with experiments on a set of Internet documents from Yahoo! that the proposed clustering algorithm can classify Web pages effectively. Keywords: Incremental update, Tree, Document, Clustering, Web, Classification 0 1 Introduction The popularity of the Internet has caused a continuous massive increase in the amount of Web pages (o... | [
901,
1711
] | Validation |
1,313 | 4 | A Reliable Multicast Protocol for Distributed Mobile Systems: Design and Evaluation Abstract Reliable multicast is a powerful communication primitive for structuring distributed programs in which multiple processes must closely cooperate together. In this paper we propose a protocol for supporting reliable multicast in a distributed system that includes mobile hosts and evaluate the performance of our proposal through simulation. We consider a scenario in which mobile hosts communicate with a wired infrastructure by means of wireless technology. Our proposal provides several novel features. The sender of each multicast may select among three increasingly strong delivery ordering guarantees: FIFO, Causal, Total. Movements do not trigger the transmission of any message in the wired network as no notion of hand-off is used. The set of senders and receivers (group) may be dynamic. Size of data structures at mobile hosts, size of message headers, number of messages in the wired network for each multicast, are all independent on the number of group members. The wireless network is assumed to provide only incomplete spatial coverage and message losses could occur even within cells. Movements are not negotiated and a mobile host that leaves a cell may enter any other cell, perhaps after a potentially long disconnection. The simulation results show that the proposed protocol has good performance and good scalability properties. 1 | [] | Test |
1,314 | 5 | Loglinear Models for First-Order Probabilistic Reasoning Recent work on loglinear models in probabilistic constraint logic programming is applied to first-order probabilistic reasoning. Probabilities are defined directly on the proofs of atomic formulae, and by marginalisation on the atomic formulae themselves. We use Stochastic Logic Programs (SLPs) composed of labelled and unlabelled definite clauses to define the proof probabilities. We have a conservative extension of first-order reasoning, so that, for example, there is a one-one mapping between logical and random variables. We show how, in this framework, Inductive Logic Programming (ILP) can be used to induce the features of a loglinear model from data. We also compare the presented framework with other approaches to first-order probabilistic reasoning. Keywords: loglinear models, constraint logic programming, inductive logic programming 1 Introduction A framework which merges first-order logical and probabilistic inference in a theoretically sound and applicable manner promises ma... | [
2378
] | Test |
1,315 | 2 | Discovering Seeds of New Interest Spread from Premature Pages Cited by Multiple Communities The World Wide Web is a great source of new topics significant for trend birth and creation. In this paper, we propose a method for discovering topics, which stimulate communities of people into earnest communications on the topics' meaning, and grow into a trend of popular interest. Here, the obtained are web pages which absorb attentions of people from multiple interest-communities. It is shown by a experiments to a small group of people, that topics in such pages can trigger the growth of peoples' interests, beyond the bounds of existing communities. | [
2459,
2503,
2986
] | Train |
1,316 | 3 | A Cost-Based Object Buffer Replacement Algorithm for Object-Oriented Database Systems Many object-oriented database systems manage object buffers to provide fast access to objects. Traditional buffer replacement algorithms based on fixed-length pages simply assume that the cost incurred by operating a buffer is proportional to the number of buffer faults. However, this assumption no longer holds in an object buffer where objects are of variable-lengths and the cost of replacing an object varies for each object. In this paper, we propose a cost-based replacement algorithm for object buffers. The proposed algorithm replaces the objects that have minimum costs per unit time and unit space. The cost model extends the previous page-based one to include the replacement costs and the sizes of objects. The performance tests show that the proposed algorithm is almost always superior to the LRU-2 algorithm and in some cases is more than twice as fast. The idea of cost-based replacement can be applied to any buffer management architectures that adopt earlier algorithms. It is espe... | [] | Test |
1,317 | 0 | An Agent-Based Approach for Manufacturing Integration - The CIIMPLEX Experience The production management system used by most manufacturers today is comprised of disconnected planning and execution processes, and lacks the support for interoperability and collaboration needed for enterprise-wide integration. This situation often prevents the manufacturer from fully exploring market opportunities in a timely fashion. To address this problem, we are exploring an agent-based approach to intelligent enterprise integration. In this approach, a set of agents with specialized expertise can be quickly assembled to help with the gathering of relevant information and knowledge, to cooperate with each other and with other parts of the production management system and humans to arrive at timely decisions in dealing with various enterprise scenarios. The proposed multi-agent system, including its architecture and implementation, are presented and demonstrated through an example integration scenario involving real planning and execution software systems. 1. Introduction The p... | [
1552
] | Train |
1,318 | 3 | A Web Odyssey: from Codd to XML INTRODUCTION The Web presents the database area with vast opportunities and commensurate challenges. Databases and the Web are organically connected at many levels. Web sites are increasingly powered by databases. Collections of linked Web pages distributed across the Internet are themselves tempting targets for a database. The emergence of XML as the lingua franca of the Web brings some much needed order and will greatly facilitate the use of database techniques to manage Web information. This paper will discuss some of the developments related to the Web from the viewpoint of database theory. As we shall see, the Web scenario requires revisiting some of the basic assumptions of the area. To be sure, database theory remains as valid as ever in the classical setting, and the database industry will continue to representamulti-billion dollar target of applicability for the foreseeable future. But the Web represents an opportunityofanentirely di#erent scale. We are th | [
29,
44,
201,
382,
910,
1228,
1298,
1575,
1600,
1612,
1646,
2215,
2594,
2910,
3141
] | Validation |
1,319 | 2 | Constrained Nearest Neighbor Queries In this paper we introduce the notion of constrained nearest neighbor queries (CNN) and propose a series of methods to answer them. This class of queries can be thought of as nearest neighbor queries with range constraints. Although both nearest neighbor and range queries have been analyzed extensively in previous literature, the implications of constrained nearest neighbor queries have not been discussed. Due to their versatility, CNN queries are suitable to a wide range of applications from GIS systems to reverse nearest neighbor queries and multimedia applications. We develop methods for answering CNN queries with different properties and advantages. We prove the optimality (with respect to I/O cost) of one of the techniques proposed in this paper. The superiority of the proposed technique is shown by a performance analysis. 1 Introduction Two dimensional range queries are used frequently in various applications such as spatial databases [Sam89, GG98] and Geographic Infor... | [
136,
575,
2090
] | Train |
1,320 | 0 | Emergent Neural Computational Architectures based on Neuroscience Present approaches for computing do not have the performance, flexibility and reliability of neural information processing systems. In order to overcome this, conventional computing systems could benefit from various characteristics of the brain such as modular organisation, robustness, timing and synchronisation, and learning and memory storage in the central nervous system. This overview incorporates some of the key research issues in the field of biologically inspired computing systems. | [
3140
] | Train |
1,321 | 2 | Context and Page Analysis for Improved Web Search NEC Research Institute has developed a metasearch engine that improves the efficiency of Web searches by downloading and analyzing each document and then displaying results that show the query terms in context. Several popular and useful search engines such as AltaVista, Excite, HotBot, Infoseek, Lycos, and Northern Light attempt to maintain full-text indexes of the World Wide Web. However, relying on a single standard search engine has limitations. The standard search engines have limited coverage, 1,2 outdated databases, and are sometimes unavailable due to problems with the network or the engine itself. The precision of standard engine results can also vary because they generally focus on handling queries quickly and use relatively simple ranking schemes. 3 Rankings can be further muddled by keyword spamming to increase a page's rank order. Often, the relevance of a particular page is obvious only after loading it and finding the query terms. Metasearch engines, such as MetaCrawler and SavvySearch, attempt to contend with the problem of limited coverage by submitting queries to several standard search engines at once. 4,5 The primary advantages of metasearch engines are that they combine the results of several search engines and present a consistent user interface. 5 However, most metasearch engines rely on the documents and summaries returned by standard search engines and so inherit their limited precision and vulnerability to keyword spamming. We developed the NEC Research Institute (NECI) metasearch engine [now called Inquirus] to improve the efficiency and precision of Web search by downloading and analyzing each document and then displaying results that show the query terms in | [
124,
130,
378,
431,
477,
505,
1108,
1352,
1529,
1767,
1883,
2503,
2532,
2558,
2569,
2627,
2822,
3037,
3170
] | Train |
1,322 | 2 | Tailoring the Interaction with Users in Web Stores . We describe the user modeling and personalization techniques adopted in SETA, a prototype toolkit for the construction of adaptive Web stores which customize the interaction with users. The Web stores created using SETA suggest the items best fitting the customers' needs and adapt the layout and the description of the store catalog to their preferences and expertise. SETA uses stereotypical information to handle the user models and applies personalization rules to dynamically generate the hypertextual pages presenting products. The system adapts the graphical aspect, length and terminology used in the descriptions to parameters like the user's receptivity, expertise and interests. Moreover, it maintains a model associated with each person the goods are selected for; in this way, multiple criteria can be applied for tailoring the selection of items to the preferences of their beneficiaries. Keywords: user modeling, personalized information presentation, customization of Web stores, ... | [
3025
] | Test |
1,323 | 0 | A Multi-Agent System for Emergent Process Management A multi-agent system manages emergent business processes. The agents in this system all have the same generic architecture. The generic agent architecture is a three-layer BDI, hybrid, multi-agent architecture. The architecture copes with plans whose goals develop and mutate. The agents in the system choose their course of action on the basis of estimates of the likelihood of a choice leading to success, and on estimates of the time, cost and value of making a choice. 1 Introduction Emergent processes are business processes; they are distinct from production workflows [1]. Emergent processes are opportunistic in nature whereas production workflows are routine. Emergent processes are inherently distributed and involve asynchronous parallel work. What amounts to a "satisfactory conclusion" of an emergent process is not generally known until the process is well advanced. Further, the tasks involved in an emergent process are typically not predefined and emerge as the process deve... | [
2364
] | Train |
1,324 | 1 | Goal Directed Adaptive Behavior in Second-Order Neural Networks: Leaning and Evolving in the MAXSON architecture The paper presents a neural network architecture (MAXSON) based on second-order connections that can learn a multiple goal approach/avoid task using reinforcement from the environment. It also enables an agent to learn vicariously, from the successes and failures of other agents. The paper shows that MAXSON can learn certain spatial navigation tasks much faster than traditional Q-learning, as well as learn goal directed behavior, increasing the agent's chances of long-term survival. The paper shows that an extension of MAXSON (V-MAXSON) enables agents to learn vicariously, and this improves the overall survivability of the agent population. | [
129,
1494,
2384
] | Train |
1,325 | 3 | Executing Suspended Logic Programs . We present an extension of Logic Programming (LP) which, in addition to ordinary LP clauses, also includes integrity constraints, explicit representation of disjunction in the bodies of clauses and in goals, and suspension of atoms as in concurrent logic languages. The resulting framework aims to unify Constraint Logic Programming (CLP), Abductive Logic Programming (ALP) and Semantic Query Optimisation (SQO) in deductive databases. We present a proof procedure for the new framework, simplifying and generalising previously proposed proof procedures for ALP. We discuss applications of the framework, formulating traditional problems from LP, ALP, CLP and SQO. Keywords: Logic Programming (LP), Constraint Logic Programming (CLP), Abductive Logic Programming (ALP), Semantic Query Optimisation (SQO) in Deductive Databases. The second author is supported by the EPSRC project "Logic-based multi-agent systems". The third author is supported by ONR grant N00014-96-1-1057. The authors are grat... | [
872,
1168,
1192,
1625,
2678
] | Test |
1,326 | 2 | Restricted Bayes Optimal Classifiers We introduce the notion of restricted Bayes optimal classifiers. These classifiers attempt to combine the flexibility of the generative approach to classification with the high accuracy associated with discriminative learning. They first create a model of the joint distribution over class labels and features. Instead of choosing the decision boundary induced directly from the model, they restrict the allowable types of decision boundaries and learn the one that minimizes the probability of misclassification relative to the estimated joint distribution. In this paper, we investigate two particular instantiations of this approach. The first uses a non-parametric density estimator — Parzen Windows with Gaussian kernels — and hyperplane decision boundaries. We show that the resulting classifier is asymptotically equivalent to a maximal margin hyperplane classifier, a highly successful discriminative classifier. We therefore provide an alternative justification for maximal margin hyperplane classifiers. The second instantiation uses a mixture of Gaussians as the estimated density; in experiments on real-world data, we show that this approach allows data with missing values to be handled in a principled manner, leading to improved performance over regular discriminative approaches. | [
988
] | Train |
1,327 | 4 | The Infocockpit: Providing Location and Place to Aid Human Memory Our work focuses on building and evaluating computer system interfaces that make information memorable. Psychology research tells us people remember spatially distributed information based on its location relative to their body, as well as the environment in which the information was learned. We apply these principles in the implementation of a multimodal prototype system, the Infocockpit (for "Information Cockpit"). The Infocockpit not only uses multiple monitors surrounding the user to engage human memory for location, but also provides ambient visual and auditory displays to engage human memory for place. We report a user study demonstrating a 56% increase in memory for information presented with our Infocockpit system as compared to a standard desktop system. | [
1472
] | Train |
1,328 | 1 | Discovering Fuzzy Classification Rules with Genetic Programming and Co-Evolution In essence, data mining consists of extracting knowledge from data. This paper proposes a co-evolutionary system for discovering fuzzy classification rules. The system uses two evolutionary algorithms: a genetic programming (GP) algorithm evolving a population of fuzzy rule sets and a simple evolutionary algorithm evolving a population of membership function definitions. The two populations co-evolve, so that the final result of the coevolutionary process is a fuzzy rule set and a set of membership function definitions which are well adapted to each other. In addition, our system also has some innovative ideas with respect to the encoding of GP individuals representing rule sets. The basic idea is that our individual encoding scheme incorporates several syntactical restrictions that facilitate the handling of rule sets in disjunctive normal form. We have also adapted GP operators to better work with the proposed individual encoding scheme. | [
1333,
2852
] | Train |
1,329 | 0 | Architecture for Agent Programming Languages . As the field of agent-based systems continues to expand rapidly, one of the most significant problems lies in being able to compare and evaluate the relative benefits and disadvantages of different systems. In part, this is due to the various different ways in which these systems are presented. One solution is to develop a set of architectural building blocks that can be used as a basis for further construction (to avoid re-inventing wheels), and to ensure a strong and effective, yet simple and accessible, means of presentation that allows for comparison and analysis of agent systems. In this paper, we address this issue in providing just such an architectural framework by using the 3APL agent programming language as a starting point for identification and specification of more general individual agent components. This provides three additional benefits: it moves the work further down the road of implementation, contributes to a growing library of agent techniques and features, and allows a detailed comparison of different agent-based systems specified in similar ways. 1 | [
1149,
2819
] | Train |
1,330 | 5 | Subpixel Localization and Uncertainty Estimation Using Occupancy Grids We describe techniques for performing mobile robot localization using occupancy grids that allow subpixel localization and uncertainty estimation in the pixelized pose space. The techniques are based on a localization method where matching is performed between the visible landmarks at the current robot position and a previously generated map of the environment. A likelihood function over the space of possible robot positions is formulated as a function of the probability distribution for the map matching error. Subpixel localization and uncertainty estimation are performed by fitting the likelihood function with a parameterized surface. The performance of the method is analyzed using synthetic experiments and an example is given using the Rocky 7 Mars rover prototype. 1 Introduction Localization is a critical issue in mobile robotics. If the robot does not know where it is, it cannot effectively plan movements, locate objects, or reach goals. It is important to not only perform accur... | [
2674
] | Validation |
1,331 | 4 | Slow Technology - Designing for Reflection Ascomputex are increcrxUkD wove intothe fabric ofexC[[OP life inteOPxC( dete mayhave tochange -- from creCOUC only fast and e ficiej tools tobe use during alimite time inspej(O situations, to crexPk1 tek1[OxC that surrounds us and the eee is a part of our activitie for longpegxUP oftime We preCUD slowtechnWOJ, : adeUD( ageD( for teP[[[Px aime at reU[[jUx andmomekj ofmekDD rek rathe thane ficieP1 inpejOP(xU[1 The aim of this pape is to dekjCC ade[j( philosophy for slowtexCPC(Cx , to discuss gecuss decus principle and to reUjjP some basic issue inintekPxUj dete from a more philosophical point ofvie . We discuss escussx of son-zz e andin;`110,HO art asinstance of slowtexCOC11x and as exkjCU[ on howthe deUP principle canbe applie inpractice Keywords : slow te xUUD[O , deU((OP human-computex inte-comput ubiquitous computing, sonitureiquitous computin 2000Springe-VexUPC Tobe publishe as: Hallns, L. &Re(OPDx J. Slow TekOOPxPP DeeOPx for ReUjD1[x Journal ofPeCkkCj and Ubiquitous Computing.Springe -VexUCj 162 1 | [
2227
] | Train |
1,332 | 4 | The Application Of Object- Oriented Analysis To Agent Based Systems Agents are important software abstractions for distributed problem solving and autonomous, pro- active behavior. They have been used in many applications, including manufacturing, enterprise integration, network management, and advanced user interfaces for worldwide web applications. As agent systems become more prevalent, the need arises for software engineering methodologies. The relationships between agents and objects, and the role of object oriented analysis in multiagent system development are discussed here. The approach is illustrated with a case study from discrete parts manufacturing. 1.0 INTRODUCTION Numerous examples can be found of applications of agent based systems to enterprise integration 31 ; concurrent engineering 5 ; and manufacturing 26 . Many agent based systems have also been developed for network management 31 , scheduling 9 , and advanced user interfaces 28 . These systems exhibit significant advances in distributed problem solving and pro-active ... | [
286,
1480,
3108
] | Train |
1,333 | 1 | A Fuzzy Beam-Search Rule Induction Algorithm . This paper proposes a fuzzy beam search rule induction algorithm for the classification task. The use of fuzzy logic and fuzzy sets not only provides us with a powerful, flexible approach to cope with uncertainty, but also allows us to express the discovered rules in a representation more intuitive and comprehensible for the user, by using linguistic terms (such as low, medium, high) rather than continuous, numeric values in rule conditions. The proposed algorithm is evaluated in two public domain data sets. 1 Introduction This paper addresses the classification task. In this task the goal is to discover a relationship between a goal attribute, whose value is to be predicted, and a set of predicting attributes. The system discovers this relationship by using known-class examples, and the discovered relationship is then used to predict the goal-attribute value (or the class) of unknown-class examples. There are numerous rule induction algorithms for the classification task. However, ... | [
1328
] | Test |
1,334 | 0 | A Plan Fusion Algorithm for Multi-Agent Systems We introduce an algorithm for cooperative planning in multi-agent systems. The algorithm enables the agents to combine (fuse) their plans in order to increase their joint profits. A computational resources and skills framework is developed for representing the planned activities of an agent under time constraints. Using this resource-skill framework, we present an ecient (polynomial time) algorithm that fuses the plans of a group of agents in such a way that their joint profits improve. The framework and the algorithm are illustrated using a simplified example from the freight transport domain. 1 Introduction Recently, much attention has been given to the topic of cooperation and cooperative planning in multiagent systems. Usually, the starting point for research on this problem is the observation that there exist classes of problems that cannot be solved by a single agent in isolation, but require several agents to work together in an interactive way, coordinating their plans and sharing ... | [
160,
3032
] | Test |
1,335 | 5 | A Methodology for Using Intelligent Agents to provide Automated Intrusion Response This paper proposes a new methodology for adaptive, automated intrusion response (IR) using software agents. The majority of intrusion response systems (IRSs) react to attacks by generating reports or alarms. This introduces a window of vulnerability between when an intrusion is detected and when action is taken to defend against the attack. Research by Cohen indicates that the success of an attack is dependent on the time gap between detection and response. If skilled attackers are given ten hours after they are detected and before a response, they will be successful 80% of the time. At thirty hours, the attacker almost never fails [1]. The proposed methodology addresses this window of vulnerability by providing an automated response to incidents using a heterogeneous collection of software agents. These agents collaborate to protect the computer system against attack and adapt their response tactics until the system administrator can take an active role in the defense of the system. | [
649,
1240,
1285
] | Train |
1,336 | 4 | BUILD-IT: an intuitive design tool based on direct object manipulation this paper, means human action in a world | [
1251
] | Test |
1,337 | 0 | Algorithm for Optimal Winner Determination in Combinatorial Auctions Combinatorial auctions, i.e. auctions where bidders can bid on com-binations of items, tend to lead to more e cient allocations than tra-ditional auctions in multi-item auctions where the agents ' valuations of the items are not additive. However, determining the winners so as to maximize revenue is NP-complete. First, existing approaches for tackling this problem are reviewed: exhaustive enumeration, dynamic programming, approximation algorithms, and restricting the allow-able combinations. Then we present our search algorithm for optimal winner determination. Experiments are shown on several bid distri-butions. The algorithm allows combinatorial auctions to scale up to signi cantly larger numbers of items and bids than prior approaches to optimal winner determination by capitalizing on the fact that the space of bids is necessarily sparsely populated in practice. The algo-rithm does this by provably su cient selective generation of children in the search tree, by using a secondary search for fast child genera-tion, by heuristics that are accurate and optimized for speed, and by four methods for preprocessing the search space. Patent pending. A highly optimized implementation of the algorithm is available for licensing both for research and commercial purposes. Please contact the author. 1 1 | [
820,
1697,
1845,
1878,
2646,
2707
] | Train |
1,338 | 3 | A Graph Query Language and Its Query Processing Many new database applications involve querying of graph data. In this paper, we present an objectoriented graph data model, and an OQL like graph query language, GOQL. The data model and the language are illustrated in the application domain of multimedia presentation graphs. We then discuss the query processing techniques for GOQL, more specifically, the translation of GOQL into an operatorbased language, called O-Algebra, extended with operators to deal with paths and sequences. We also discuss different approaches for efficient implementations of algebra operators for paths and sequences. 1 Introduction Many database applications such as hypertext applications, geographic information systems, world wide web searching, and heterogeneous information integration, etc., require modeling and querying of graph data ([Guti94, GBPV 94, MeMM 96, BDHS 96, AQMWW 96, AM 98, FFKLS 98, LSBBOO98]). In this paper, we present a data model, and an OQL-like query language GOQL, for querying graphs. ... | [
1941
] | Validation |
1,339 | 0 | Supporting Trust in Virtual Communities At any given time, the stability of a community depends on the right balance of trust and distrust. Trust is also the basis of economic activities, making such things as credit agreements, business contracts and customer confidence possible. In real-life, we are faced with increasingly complex decisions due to the increasingly daunting range of options provided by the Internet and uncertainty in the credibility of virtual entities. In short, information overload, increased uncertainty and risk taking a prominent feature of modern living. We as members of society cope with these complexities and uncertainties by relying on a vital social phenomenon, trust, which forms the basis of all social interaction. However, the ability to reason about trust is absent from the virtual medium, making virtual communities of real and artificial agents fragile. This is unsatisfactory as all `virtual' interactions are ultimately human-bound. Therefore we need a trust model to allow artificial agents to ... | [
3095
] | Validation |
1,340 | 0 | A Taxonomy of Web Agents In this paper we propose a taxonomy of web agents, which encompasses agents that provide a text-based interface to for example information retrieval services as well as avatarembodied guides that help visitors to navigate in virtual environments. Our taxonomy must be regarded as an instrument to delineate targets for research and the realization of prototype applications that demonstrate the usefulness of agent-based intelligence on the Web. In addition, we deploy our agent-taxonomy to establish the implications particular target applications have with respect to software architecture and computational resources. 1. Introduction There is a lot of interest and work in the research and development of agent technology with applications on the Web. Many types of web agents have been proposed in recent years, which range from domain-dependent agents, like e-commerce agents, information gathering agents, to function-dependent agents, like negotiation agents, cooperating agents. In addition... | [
1978
] | Train |
1,341 | 3 | Database Query Formation from Natural Language using Semantic Modeling and Statistical Keyword Meaning Disambiguation This paper describes a natural language interface to database systems which is based on the query formation capabilities of a High-level Query Formulator. The formulator relies on the Semantic Graph of the database, which is a model of the data stored in the database. The natural language interface accepts a user input in natural language and extracts the necessary information needed by the formulator. This extraction process is performed using keywords obtained from the Semantic Graph and the database. Because keywords may have several meanings within a given domain, keyword meaning disambiguation is done using a statistical approach which involves comparing vectors of n-grams. N-grams are n contiguous words within a given text of natural language and they are capable of capturing lexical context. Traditionally, natural language interfaces have been heavy with grammars and other knowledge, but have been wide-ranging in functionality. The interface presented in this paper is more porta... | [
2582
] | Train |
1,342 | 5 | Intelligent Support for Enterprise Modelling Enterprise modelling- integrating models of all pertinent aspects of an enterprise- is essential to the management of change in organisations. An integrated view of an organisation provides insight into what aspects may bechanged, how they may be changed, and what the overall e ect of speci c changes will be. AIAI at the University ofEdinburgh has an ongoing research programme which focuses on the use AI techniques to cover the requirements of enterprise modelling and the tools to support it. The AI techniques used in AIAI's programme range from knowledge representation, ontologies and process modelling techniques to visualisation techniques, intelligent work ow and coordination technology. The techniques are combined in an integrated toolset delivered on an agent-based architecture. Part of AIAI's programme is the Enterprise project which has been instrumental in determining the requirements for enterprise modelling and in the development ofan integrated toolset to support it. The results of the Enterprise project show thatwhen combined with task management support, enterprise models may directly control the operation of an organisation. Based on the results of the Enterprise project, AIAI's TBPM project currently addresses coordination issues of enterprise modelling support. In this paper, we rst describe the requirements for enterprise modelling and enactment in general. We then discuss the Enterprise Toolset which was designed and was implemented to address these requirements. Finally, weevaluate the toolset and describe extensions that are currently being undertaken. AIAI-TR-220/0.1 Page 1 1 | [
1789
] | Train |
1,343 | 3 | Transactional Coordination Agents for Composite Systems Composite systems are collections of autonomous, heterogeneous, and distributed software applications. In these systems, data dependencies are continuously violated by local operations and therefore, coordination processes are necessary to guarantee overall correctness and consistency. Such coordination processes must be endowed with some form of execution guarantees, which require the participating subsystems to have certain database functionality (such as atomicity of local operations, order-preservation and either compensation of operations or the deferment of their commit). However, this functionality is not present in many applications and must be implemented by a transactional coordination agent coupled with the application. In this paper, we discuss the requirements to be met by the applications and their associated transactional coordination agents. We identify a minimal set of functionality the applications must provide in order to participate in transactional coordination pro... | [
285,
1534
] | Train |
1,344 | 0 | Improvisational Synthetic Actors with Flexible Personalities We provide synthetic agents as intelligent actors that can improvise their behaviors in interactive environments without detailed planning, just as human improvisers do. Their behavior is based on the directions they receive: high-level scenarios, user commands, and personalities of the characters portrayed. We propose a social-psychological model that enables an author to define a character's personality influenced by moods and interpersonal relationships. Using examples of characters and experiments with users, we show how such a model can be exploited by synthetic actors to produce performances that are theatrically interesting, believable, and diverse. Keywords: synthetic actors, personality, improvisation, believability, entertainment. 1. Introduction Personality is the set of psychological traits that distinguish an individual from all others and characterize his or her behavior (Hayes-Roth et al. 1996). Such traits can easily be recognized by others, and people are commonly ... | [
2247
] | Train |
1,345 | 1 | An Introduction to Variational Methods for Graphical Methods . This paper presents a tutorial introduction to the use of variational methods for inference and learning in graphical models (Bayesian networks and Markov random fields). We present a number of examples of graphical models, including the QMR-DT database, the sigmoid belief network, the Boltzmann machine, and several variants of hidden Markov models, in which it is infeasible to run exact inference algorithms. We then introduce variational methods, which exploit laws of large numbers to transform the original graphical model into a simplified graphical model in which inference is efficient. Inference in the simpified model provides bounds on probabilities of interest in the original model. We describe a general framework for generating variational transformations based on convex duality. Finally we return to the examples and demonstrate how variational algorithms can be formulated in each case. | [
1126,
1887,
2463
] | Test |
1,346 | 0 | eMediator: a Next Generation Electronic Commerce Server This paper presents eMediator, an electronic commerce server prototype that demonstrates ways in which algorithmic support and game-theoretic incentive engineering can jointly improve the efficiency of ecommerce. eAuctionHouse, the configurable auction server, includes a variety of generalized combinatorial auctions and exchanges, pricing schemes, bidding languages, mobile agents, and user support for choosing an auction type. We introduce two new logical bidding languages for combinatorial markets: the XOR bidding language and the OR-of-XORs bidding language. Unlike the traditional OR bidding language, these are fully expressive. They therefore enable the use of the Clarke-Groves pricing mechanism for motivating the bidders to bid truthfully. eAuctionHouse also supports supply/demand curve bidding. eCommitter, the leveled commitment contract optimizer, determines the optimal contract price and decommitting penalties for a variety of leveled commitment contracting mechanisms, taking into account that rational agents will decommit strategically in Nash equilibrium. It also determines the optimal decommitting strategies for any given leveled commitment contract. eExchangeHouse, the safe exchange planner, enables unenforced anonymous exchanges by dividing the exchange into chunks and sequencing those chunks to be delivered safely in alternation between the buyer and the seller. | [
820,
1680
] | Validation |
1,347 | 5 | Three New Algorithms for Projective Bundle Adjustment with minimal parameters Bundle adjustment is a technique used to compute the maximum likelihood estimate of structure and motion from image feature correspondences. It practice, large non-linear systems have to be solved, most of the time using an iterative optimization process starting from a sub-optimal solution obtained by using linear methods. The behaviour, in terms of convergence, and the computational cost of this process depend on the parameterization used to represent the problem, i.e. of structure and motion. | [
1417
] | Train |
1,348 | 0 | Programming Satan's Agents Mobile agent security is still a young discipline and most naturally, the focus up to the time of writing was on inventing new cryptographic protocols for securing various aspects of mobile agents. However, past experience shows that protocols can be flawed, and flaws in protocols can remain unnoticed for a long period of time. The game of breaking and fixing protocols is a necessary evolutionary process that leads to a better understanding of the underlying problems and ultimately to more robust and secure systems. Although, to the best of our knowledge, little work has been published on breaking protocols for mobile agents, it is inconceivable that the multitude of protocols proposed so far are all flawless. As it turns out, the opposite is true. We identify flaws in protocols proposed by Corradi et al., Karjoth et al., and Karnik et al., including protocols based on secure coprocessors. | [
546,
2715,
2843
] | Train |
1,349 | 4 | Extracting Low Resolution Text with an Active Camera for OCR Reading text in any scene is useful in the context of wearable computing, robotic vision or as an aid for visually handicapped people. Here, we present a novel automatic text reading system using an active camera focused on text regions already located in the scene (using our recent work). A region of text found is analysed to determine the optimal zoom that would foveate onto it. Then a number of images are captured over the text region to reconstruct a high-resolution mosaic of the whole region. This magnified image of the text is good enough for reading byhumans or for recognition by OCR. Even with a low resolution camera we obtained very good results. | [
928
] | Validation |
1,350 | 1 | Layered Learning This paper presents layered learning, a hierarchical machine learning paradigm. Layered learning applies to tasks for which learning a direct mapping from inputs to outputs is intractable with existing learning algorithms. Given a hierarchical task decomposition into subtasks, layered learning seamlessly integrates separate learning at each subtask layer. The learning of each subtask directly facilitates the learning of the next higher subtask layer by determining at least one of three of its components: (i) the set of training examples; (ii) the input representation; and/or (iii) the output representation. We introduce layered learning in its domainindependent general form. We then present a full implementation in a complex domain, namely simulated robotic soccer. 1. Introduction Machine learning (ML) algorithms select a hypothesis from a hypothesis space based on a set of training examples such that the chosen hypothesis is predicted to characterize unseen examples... | [
2388,
2614
] | Train |
1,351 | 3 | Static and Dynamic Information Organization with Star Clusters In this paper we present a system for static and dynamic information organization and show our evaluations of this system on TREC data. We introduce the off-line and on-line star clustering algorithms for information organization. Our evaluation experiments show that the off-line star algorithm outperforms the single link and average link clustering algorithms. Since the star algorithm is also highly efficient and simple to implement, we advocate its use for tasks that require clustering, such as information organization, browsing, filtering, routing, topic tracking, and new topic detection. | [
2586
] | Test |
1,352 | 2 | Improving Category Specific Web Search by Learning Query Modifications A user searching for documents within a specific category using a general purpose search engine might have a difficult time finding valuable documents. To improve category specific search, we show that a trained classifier can recognize pages of a specified category with high precision by using textual content, text location, and HTML structure. We show that query modifications to web search engines increase the probability that the documents returned are of the specific category. We evaluate the effectiveness of several query modifications on real search engines, showing that the approach is highly effective for locating personal homepages and calls for papers. 1: Introduction Typical web search engines index millions of pages across a variety of categories, and return results ranked by expected topical relevance. Only a small percentage of these pages may be of a specific category, for example, personal homepages, or calls for papers. A user may examine large numbers of pages abou... | [
130,
1108,
1136,
1321,
2178,
2188,
2569,
2822,
3077
] | Train |
1,353 | 2 | PeerWare: Core Middleware Support for Peer-to-Peer and Mobile Systems The pervasiveness of computer networks, together with the availability of wireless links, are steering distributed systems towards scenarios where computing is increasingly decentralized, decoupled, and dynamically reconfigurable. The popularity of and demand for applications that exploit mobile and peer-to-peer interactions is a symptom of such change. Nevertheless, by and large these applications are being built in an ad hoc manner, and often with architectures that, by sticking to the traditional client-server paradigm, do not fully capture and support the peculiar requirements of the new scenario. | [
580
] | Validation |
1,354 | 1 | Genetic Algorithms for Binary Quadratic Programming In this paper, genetic algorithms for the unconstrained binary quadratic programming problem (BQP) are presented. It is shown that for small problems a simple genetic algorithm with uniform crossover is sufficient to find optimum or best-known solutions in short time, while for problems with a high number of variables (n 200) it is essential to incorporate local search to arrive at high-quality solutions. A hybrid genetic algorithm incorporating local search is tested on 40 problem instances of sizes containing between n = 200 and n = 2500. The results of the computer experiments show that the approach is comparable to alternative heuristics such as tabu search for small instances and superior to tabu search and simulated annealing for large instances. New best solutions could be found for 14 large problem instances. 1 INTRODUCTION In the unconstrained binary quadratic programming problem (BQP), a symmetric rational n \Theta n matrix Q = (q ij ) is given, and a binary vector of leng... | [
2755
] | Test |
1,355 | 3 | AL-log: Integrating Datalog and Description Logics . We presentanintegrated system for knowledge representation, called AL-log, based on description logics and the deductive database language Datalog. AL-log embodies two subsystems, called structural and relational. The former allows for the de#nition of structural knowledge about classes of interest #concepts# and membership relation between objects and classes. The latter allows for the de#nition of relational knowledge about objects described in the structural component. The interaction between the two components is obtained by allowing constraints within Datalog clauses, thus requiring the variables in the clauses to range over the set of instances of a speci#ed concept. We propose a method for query answering in AL-log based on constrained resolution, where the usual deduction procedure de#ned for Datalog is integrated with a method for reasoning on the structural knowledge. Keywords: Description Logics, Deductive Databases, Datalog, Object-based Knowledge Representation, Query A... | [
2990
] | Train |
1,356 | 3 | The String B-Tree: A New Data Structure for String Search in External Memory and its Applications. We introduce a new text-indexing data structure, the String B-Tree, that can be seen as a link between some traditional external-memory and string-matching data structures. In a short phrase, it is a combination of B-trees and Patricia tries for internal-node indices that is made more effective by adding extra pointers to speed up search and update operations. Consequently, the String B-Tree overcomes the theoretical limitations of inverted files, B-trees, prefix B-trees, suffix arrays, compacted tries and suffix trees. String B-trees have the same worst-case performance as B-trees but they manage unbounded-length strings and perform much more powerful search operations such as the ones supported by suffix trees. String B-trees are also effective in main memory (RAM model) because they improve the online suffix tree search on a dynamic set of strings. They also can be successfully applied to database indexing and software duplication. | [
2383
] | Train |
1,357 | 2 | FEATURES: Real-time Adaptive Feature Learning and Document Learning for Web Search In this paper we report our research on building Features - an intelligent web search engine that is able to perform real-time adaptive feature (i.e., keyword) and document learning. Not only does Features learn from the user's document relevance feedback, but also automatically extracts and suggests indexing keywords relevant to a search query and learns from the user's keyword relevance feedback so that it is able to speed up its search process and to enhance its search performance. We design two efficient and mutual-benefiting learning algorithms that work concurrently, one for feature learning and the other for document learning. Features employs these algorithms together with an internal index database and a real-time meta-searcher so to perform adaptive real-time learning to find desired documents with as little relevance feedback from the user as possible. The architecture and performance of Features are also discussed. 1 Introduction As the world wide web rapidly evo... | [
95,
147,
166,
1016,
1269,
2459,
2475,
2503
] | Train |
1,358 | 0 | Social Mental Shaping: Modelling the Impact of Sociality on the Mental States of Autonomous Agents This paper presents a framework that captures how the social nature of agents that are situated in a multi-agent environment impacts upon their individual mental states. Roles and social relationships provide an abstraction upon which we develop the notion of social mental shaping. This allows us to extend the standard Belief-DesireIntention model to account for how common social phenomena (e.g. cooperation, collaborative problem-solving and negotiation) can be integrated into a unified theoretical perspective that reflects a fully explicated model of the autonomous agent's mental state. Keywords: Multi-agent systems, agent interactions, BDI models, social influence. 3 1. | [
313,
1260,
1724
] | Train |
1,359 | 1 | The Pareto Archived Evolution Strategy : A New Baseline Algorithm for Pareto Multiobjective Optimisation Most popular evolutionary algorithms for multiobjective optimisation maintain a population of solutions from which individuals are selected for reproduction. In this paper, we introduce a simpler evolution scheme for multiobjective problems, called the Pareto Archived Evolution Strategy (PAES). We argue that PAES may represent the simplest possible non-trivial algorithm capable of generating diverse solutions in the Pareto optimal set. The algorithm is identified as being a (1 + 1) evolution strategy, using local search from a population of one but using a reference archive of previously found solutions in order to identify the approximate dominance ranking of the current and candidate solution vectors. PAES is intended as a good baseline approach, against which more involved methods may be compared, and may also serve well in some real-world applications when local search seems superior to or competitive with population-based methods. The performance of the new algorithm is compared w... | [
652,
2061
] | Train |
1,360 | 1 | Text Categorization Using Weight Adjusted k-Nearest Neighbor Classification . Automatic text categorization is an important task that can help people finding information on huge online resources. Text categorization presents unique challenges due to the large number of attributes present in the data set, large number of training samples, attribute dependency, and multi-modality of categories. Existing classification techniques have limited applicability in the data sets of these natures. In this paper, we present a Weight Adjusted k-Nearest Neighbor (WAKNN) classification that learns feature weights based on a greedy hill climbing technique. We also present two performance optimizations of WAKNN that improve the computational performance by a few orders of magnitude, but do not compromise on the classification quality. We experimentally evaluated WAKNN on 52 document data sets from a variety of domains and compared its performance against several classification algorithms, such as C4.5, RIPPER, Naive-Bayesian, PEBLS and VSM. Experimental results ... | [
495,
630,
737,
1405,
1796,
2780,
2794
] | Train |
1,361 | 2 | Mining the Web to Create Minority Language Corpora The Web is a valuable source of language specific resources but the process of collecting, organizing and utilizing these resources is difficult. We describe CorpusBuilder, an approach for automatically generating Web-search queries for collecting documents in a minority language. It differs from pseudo-relevance feedback in that retrieved documents are labeled by an automatic language classifier as relevant or irrelevant, and this feedback is used to generate new queries. We experiment with various query-generation methods and query-lengths to find inclusion/exclusion terms that are helpful for retrieving documents in the target language and find that using odds-ratio scores calculated over the documents acquired so far was one of the most consistently accurate query-generation methods. We also describe experiments using a handful of words elicited from a user instead of initial documents and show that the methods perform similarly. Experiments applying the same approach to multiple languages are also presented showing that our approach generalizes to a variety of languages. 1. | [
95,
521,
1140,
1234,
2580
] | Test |
1,362 | 4 | CollabLogger: A Tool for Visualizing Groups At Work The CollabLogger is a visual tool that supports usability analyses of human-computer interaction in a team environment. Participants in our computer-mediated activity were engaged in a small-scale manufacturing testbed project. Interactions of the group were mediated by Teamwave Workplace 1 and the members performed both synchronous and asynchronous activities depending on their availability, project requirements, and due to chance meetings in the collaborative space. The software was instrumented to log users' interactions with the system and each other. The CollabLogger addresses the problem of helping investigators analyze the volumes of log data that groupware tools can generate. Visual tools are powerful when large amounts of diverse data present themselves. The place-based collaboration environment offered by Teamwave Workplace provided a level of organization that allowed us to create a visual interface with which to perform exploratory sequential data analysis. Preliminary ... | [
99,
849,
2076
] | Test |
1,363 | 2 | Experience with a Language Engineering Architecture: Three Years of GATE GATE, the General Architecture for Text Engineering, aims to provide a software infrastructure for researchers and developers working in the area of natural language processing. A version of GATE has now been widely available for three years. In this paper we review the objectives which motivated the creation of GATE and the functionality and design of the current system. We discuss the strengths and weaknesses of the current system, identify areas for improvement and present plans for implementing these improvements. Introduction We think that if you're researching human language processing you should probably not be writing code to: ffl store data on disk; ffl display data; ffl load processor modules and data stores into processes; ffl initiate and administer processes; ffl divide computation between client and server; ffl pass data between processes and machines. A Domain-Specific Software Architecture (DSSA) for language processing should do all this for you. You will have t... | [
1295
] | Test |
1,364 | 4 | Simulated 3D Painting This technical report looks at the motivation for simulating painting directly on 3D objects, and investigates the main issues faced by such systems. These issues include the provision of natural user interfaces, the reproduction of realistic brush effects and the surface parameterization for texture mapping so that the results of the painting can be stored on texture maps. The technical report further investigates the issues involved in using a haptic interface for simulating 3D painting, and the issues in surface parameterization for texture mapping with application to 3D painting. A survey of some work related to 3D painting, haptic rendering and surface parameterization for texture mapping is presented. 1 | [
1148,
1633
] | Validation |
1,365 | 0 | Multi-Agent Systems by Incremental Gradient Reinforcement Learning situated with local and scalable perceptions, . have identical capabilities, . are possibly heterogeneous, . cooperate, . do not directly communicate. Each agent learns its behavior on its own. 11/22 IJCAI'01 # # # # # # Bloc merging (the problem) . reward: +3 if blocs are merged . actions: N W E S . perceptions: -- dir(agent) #4 -- dir(yellow bloc) -- dir(blue bloc) -- near(yellow bloc) -- near(blue bloc) total 1024/4 (MDP with 2 agents and 2 cubes for an 8 8 world: 15.248.024 states !!!) 12/22 IJCAI'01 # # # # # # An agent learns a policy: # : But the < O, A > is not markovian: . convergence is not assured, . stochastic policies should perform better [SJJ94]. 1b 1a A (-R) B(+R) B(-R) A(+R) 13/22 IJCAI'01 # # # # # # Multi-agent framework: . each agent considers other agents as part of the environment, . all agents learn, therefore evolve, unpredictable transitions. Q-learning Baxter's gradient descent A | [
1487,
3012
] | Train |
1,366 | 2 | Enhancing Supervised Learning with Unlabeled Data In many practical learning scenarios, there is a small amount of labeled data along with a large pool of unlabeled data. Many supervised learning algorithms have been developed and extensively studied. We present a new "co-training" strategy for using unlabeled data to improve the performance of standard supervised learning algorithms. Unlike much of the prior work, such as the co-training procedure of Blum and Mitchell (1998), we do not assume there are two redundant views both of which are sufficient for perfect classification. The only requirement our co-training strategy places on each supervised learning algorithm is that its hypothesis partitions the example space into a set of equivalence classes (e.g. for a decision tree each leaf defines an equivalence class). We evaluate our co-training strategy via experiments using data from the UCI repository. 1. Introduction In many practical learning scenarios, there is a small amount of labeled data along with a lar... | [
855,
1386,
1478,
2176
] | Train |
1,367 | 0 | Spatial Agents Implemented in a Logical Expressible Language In this paper, we present a multi-layered architecture for spatial and temporal agents. The focus is laid on the declarativity of the approach, which makes agent scripts expressive and well understandable. They can be realized as (constraint) logic programs. The logical description language is able to express actions or plans for one and more autonomous and cooperating agents for the RoboCup (Simulator League). The system architecture hosts constraint technology for qualitative spatial reasoning, but quantitative data is taken into account, too. The basic (hardware) layer processes the agent's sensor information. An interface transfers this lowlevel data into a logical representation. It provides facilities to access the preprocessed data and supplies several basic skills. The second layer performs (qualitative) spatial reasoning. On top of this, the third layer enables more complex skills such as passing, offside-detection etc. At last, the fourth layer establishes acting as a team both by emergent and explicit cooperation. Logic and deduction provide a clean means to specify and also to implement teamwork behavior. 1 | [
403,
780,
1911,
2182
] | Train |
1,368 | 2 | Efficient Learning of Semi-structured Data from Queries This paper studies the polynomial-time learnability of the classes of ordered gapped tree patterns (OGT) and ordered gapped forests (OGF) under the into-matching semantics in the query learning model of Angluin. The class OGT is a model of semi-structured database query languages, and a generalization of both the class of ordered/unordered tree pattern languages and the class of non-erasing regular pattern languages. | [
3099
] | Train |
1,369 | 0 | Mansion: A Structured Middleware Environment for Agents Developing processes intended to roam in largescale heterogeneous distributed systems is difficult: their environment is unstructured, and interoperability issues often emerge. Mansion is a new paradigm to provide a large-scale structured environment for mobile agents (autonomous migrating processes). Security, transparency, and interoperability are important design guidelines for the development of Mansion. | [
2313,
2843
] | Test |
1,370 | 0 | Agents Acting and Moving in Healthcare Scenario: A Paradigm for Telemedical Collaboration The present paper describes a novel approach to the analysis and development of telemedicine systems, based on the multi-agent paradigm. An agent is an autonomous, social, reactive and proactive entity, sometimes also mobile. Since telemedicine is grounded on communication and sharing of resources, agents are suitable for its analysis and implementation, and we adopted them for developing a prototype telemedical agent. | [
2983
] | Train |
1,371 | 5 | A Parametric Alternative to Grids for Occupancy-Based World Modeling In the paper, we consider an occupancy-based approach for range data fusion, as it is used in mobile robotics. We tackle the major problem of this approach, which is the redundancy of stored and processed data caused by using the grid representation of the occupancy function, by proposing a parametric piece-wise linear representation. When applied to the vision-based world exploration, the new representation is shown to have advantages over the former one, which include its suitability for radial range data, its efficiency in representing and fusing range data, and its convenience for navigation map extraction. The proposed technique is implemented on a mobile robot, Boticelli. The results obtained from running the robot are presented. 1 Introduction In mobile robot world exploration, the occupancybased approach is one of the most commonly used [7, 13, 3, 2, 8, 4, 11]. In this approach, the exploration policy is determined by the occupancy model of the world which is built from the r... | [
2577
] | Train |
1,372 | 2 | The Cambridge University Spoken Document Retrieval System This paper describes the spoken document retrieval system that we have been developing and assesses its performance using automatic transcriptions of about 50 hours of broadcast news data. The recognition engine is based on the HTK broadcast news transcription system and the retrieval engine is based on the techniques developed at City University. The retrieval performance over a wide range of speech transcription error rates is presented and a number of recognition error metrics that more accurately reflect the impact of transcription errors on retrieval accuracy are defined and computed. The results demonstrate the importance of high accuracy automatic transcription. The final system is currently being evaluated on the 1998 TREC-7 spoken document retrieval task. 1. | [
1070,
1792,
1895
] | Validation |
1,373 | 3 | Implementation and Performance Analysis of Incremental Equations of Nested Relations View materialization is an important way of improving the performance of query processing. When an update occurs to the source data from which a materialized view is derived, the materialized view has to be updated so that it is consistent with the source data. This update process is called view maintenance. The incremental method of view maintenance, which computes the new view using the old view and the update to the source data, is widely preferred to full view recomputation when the update is small in size. The small update size becomes an important concept for measuring the cheap performance of the incremental methods. In this paper, we investigate what is the limit of the small update size, which we call size limit for the incremental maintenance. When the size of an update exceeds the limit, the incremental maintenance is no longer cheaper than the view recomputation. The investigation is based on incremental equations for operators in the nested relational model... | [
1576,
2507
] | Train |
1,374 | 2 | ACIRD: Intelligent Internet Documents Organization and Retrieval This paper presents an intelligent Internet information system, Automatic Classifier for the Internet Resource Discovery (ACIRD), which uses machine learning techniques to organize and retrieve Internet documents. ACIRD consists of a knowledge acquisition process, document classifier and two-phase search engine. The knowledge acquisition process of ACIRD automatically learns classification knowledge from classified Internet documents. The document classifier applies learned classification knowledge to classify newly collected Internet documents into one or more classes. Experimental results indicate that ACIRD performs as well or better than human experts in both knowledge acquisition and document classification. By using the learned classification knowledge and the given class lattice, the ACIRD two-phase search engine responds to user queries with hierarchically structured navigable results (instead of a conventional flat ranked document list), which greatly aids users in locating information from numerous, diversified Internet documents. | [
2779
] | Train |
1,375 | 2 | Automatic Text Detection and Tracking in Digital Video Text which either appears in a scene or is graphically added to video can provide an important supplemental source of index information as well as clues for decoding the video's structure and for classification. In this paper we present algorithms for detecting and tracking text components that appear within digital video frames. Our system implements a scale-space feature extractor that feeds an artificial neural processor to extract textual regions and track their movement over time. The extracted regions can then be used as input to an appropriate Optical Character Recognition system which produces indexible keywords. Keywords: Text Detection, Text Tracking, Video Indexing, Digital Libraries, Neural Network, Wavelet The support of this effort by the Department of Defense under contract MDA 9049-6C-1250 is gratefully acknowledged. 1 Introduction The increasing availability of online digital imagery and video has rekindled interest in the problems of how to index multimedia informa... | [
17,
2018,
2759,
2938
] | Validation |
1,376 | 2 | Locality in Search Engine Queries and Its Implications for Caching Caching is a popular technique for reducing both server load and user response time in distributed systems. In this paper, we are interested in the question of whether caching might be effective for search engines as well. We study two real search engine traces by examining query locality and its implications for caching. Our trace analysis results show that: (1) Queries have significant locality, with query frequency following a Zipf distribution. Very popular queries are shared among different users and can be cached at servers or proxies, while 16 % to 22% of the queries are from the same users and should be cached at the user side. Multiple-word queries are shared less and should be cached mainly at the user side. (2) If caching is to be done at the user side, short-term caching for hours will be enough to cover query temporal locality, while server/proxy caching should be based on longer periods such as days. (3) Most users have small lexicons when submitting queries. Frequent users who submit many search requests tend to reuse a small subset of words to form queries. Thus, with proxy or user side caching, prefetching based on user lexicon looks promising. | [
711
] | Train |
1,377 | 2 | Evolving Personal Agent Environments To Reduce Internet Information Overload: Initial Considerations this paper we will address the other major information overload problem, arising from habitual surfing. We will present our proposed research programme and also outline and support a potential solution. So far this issue has been far less visited and researched, though with the ever increasing use of computers at home and in the workplace, as well as the imminent arrival of information appliances, this may well change [6]. | [
676,
1857,
2336
] | Train |
1,378 | 1 | Learning Situation-Specific Control In Multi-Agent Systems The work presented in this thesis deals with techniques to improve problem solving control skills of cooperative agents through machine learning. In a multi-agent system, the local problem solving control of an agent can interact in complex and intricate ways with the problem solving control of other agents. In such systems, an agent cannot make effective control decisions based purely on its local problem solving state. Effective cooperation requires that the global problem-solving state influence the local control decisions made by an agent. We call such an influence cooperative control. An agent with a purely local view of the problem solving situation cannot learn ... | [
286,
306,
1107,
2835
] | Test |
1,379 | 3 | Partial and Complete Tuples and Sets in Deductive Databases . In a nested relational or complex object database, nested tuples and sets are used to represent real world objects. For various reasons, such tuples and sets can be partial or complete. In this paper, we discuss how to support them in deductive databases. In particular, we present a deductive database language RLOG II that supports partial and complete tuples and sets based on Relationlog. This work provides a firm logical foundation for nested relational and complex object databases that have both partial and complete tuples and sets. 1 Introduction In a nested relational database or complex object database, nested tuples and sets are used to represent real world objects. A relation is just a set of tuples and a database is a tuple of relations. For various reasons, the information about real world objects in a database may be incomplete. Hence, both tuples and sets can be partial or complete. In the past several years, a sub-problem, that is, a database which contains partial an... | [
2083
] | Train |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.