node_id int64 0 76.9k | label int64 0 39 | text stringlengths 13 124k | neighbors listlengths 0 3.32k | mask stringclasses 4
values |
|---|---|---|---|---|
2,080 | 3 | A Scalable Algorithm for Answering Queries Using Views The problem of answering queries using views is to find efficient methods of answering a query using a set of previously materialized views over the database, rather than accessing the database relations. The problem has received significant attention because of its relevance to a wide variety of data management problems, such as data integration, query optimization, and the maintenance of physical data independence. To date, the performance of proposed algorithms has received very little attention, and in particular, their scale up in the presence of a large number of views is unknown. We first analyze two previous algorithms, the bucket algorithm and the inverse-rules algorithm, and show their deficiencies. We then describe the MiniCon algorithm, a novel algorithm for finding the maximally-contained rewriting of a conjunctive query using a set of conjunctive views. We present the first experimental study of algorithms for answering queries using views. The study shows that the MiniCon algorithm scales up well and significantly outperforms the previous algorithms. Finally, we describe an extension of the MiniCon algorithm to handle comparison predicates, and show its performance experimentally. Thanks to Daniela Florescu, Marc Friedman, Zack Ives, Ioana Manolescu, Dan Weld, and Steve Wolfman for their comments on earlier drafts of this paper. This research was funded by a Sloan Fellowship, NSF Grant #IIS-9978567, a NSF Graduate Research Fellowship, and a Lucent Technologies GRPW Grant Permission to copy without fee all or part of this material is granted provided that the copies are not made or distributed for direct commercial advantage, the VLDB copyright notice and the title of the publication and its date appear, and notice is given tha... | [
40,
763,
965,
1217,
1612
] | Validation |
2,081 | 5 | Nonmonotonic Reasoning: Towards Efficient Calculi and Implementations | [
1581,
2417
] | Train |
2,082 | 4 | Gesture Recognition for Visually Mediated Interaction . This paper reports initial research on supporting Visually Mediated Interaction (VMI) by developing person-specific and generic gesture models for the control of active cameras. We describe a time-delay variant of the Radial Basis Function (TDRBF) network and evaluate its performance on recognising simple pointing and waving hand gestures in image sequences. Experimental results are presented that show that high levels of performance can be obtained for this type of gesture recognition using such techniques, both for particular individuals and across a set of individuals. Characteristic visual evidence can be automatically selected, depending on the task demands. 1 Introduction In general, robust tracking of non-rigid objects such as human bodies is difficult due to rapid motion, occlusion and ambiguities in segmentation and model matching. Ongoing research at the MIT Media Lab has shown progress in the modelling and interpretation of human body activity [24, 30, 31]. Compu... | [
1963
] | Train |
2,083 | 3 | Relationlog: A Typed Extension to Datalog with Sets and Tuples This paper presents a novel logic programming based language for nested relational and complex value models called Relationlog. It stands in the same relationship to the nested relational and complex value models as Datalog stands to the relational model. The main novelty of the language is the introduction of powerful mechanisms, namely, partial and complete set terms, for representing and manipulating both partial and complete information on nested sets, tuples and relations. They generalize the set grouping and set enumeration mechanisms of LDL and allow the user to directly encode the open and closed world assumptions on nested sets, tuples, and relations. They allow direct inference and access to deeply embedded values in a complex value relation as if the relation is normalized, which greatly increases the ease of use of the language. As a result, the extended relational algebra operations can be represented in Relationlog directly, and more importantly, recursively in a way similar to Datalog. Like Datalog, Relationlog has a well-defined Herbrand model-theoretic semantics, which captures the intended semantics of nested sets, tuples and relations, and also a well-defined proof-theoretic semantics which coincides with its model-theoretic semantics. | [
418,
761,
821,
1116,
1379,
1482,
1940,
2277,
2588,
2618,
2832
] | Test |
2,084 | 3 | Compiling Standard ML to Java Bytecodes ing with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept, ACM Inc., fax +1 (212) 869-0481, or permissions@acm.org. Compiling Standard ML to Java Bytecodes Nick Benton Andrew Kennedy George Russell Persimmon IT, Inc. Cambridge, U.K. fnick,andrew,georgeg@persimmon.co.uk Abstract MLJ compiles SML'97 into verifier-compliant Java bytecodes. Its features include type-checked interlanguage working extensions which allow ML and Java code to call each other, automatic recompilation management, compact compiled code and runtime performance which, using a `just in time' compiling Java virtual machine, usually exceeds that of existing specialised bytecode interpreters for ML. Notable features of the compiler itself include whole-program optimisation based on rewriting, compilation of polymorphism by specialisation, a novel monadic intermediate lang... | [
1037
] | Train |
2,085 | 3 | Distributed Safety Controllers for Web Services We show how to use high-level synchronization constraints, written in a version of monadic second-order logic on finite strings, to synthesize safety controllers for interactive web services. We improve on the naïve runtime model to avoid state-space explosions and to increase the flow capacities of services. | [
328
] | Validation |
2,086 | 4 | The Importance of Being Mobile: Some Social Consequences of Wearable Augmented Reality Systems What are the consequences of mobility for augmented reality ? This brief paper explores some of the issues that I believe will be raised by the development and future commonplace adoption of mobile, wearable, augmented reality systems. These include: social influences on tracking accuracy, the importance of appearance and comfort, an increase in collaborative applications, integration with other devices, and implications for personal privacy. 1. Introduction Over the past decade, the reality of mobile computing has begun to embrace the potential of wearable computing. In the process, several researchers have attempted to clarify what distinguishes wearable computing from mobile computing. Rhodes [10] suggests five criteria for wearable systems: portable while operational, needing minimal manual input, sensitive to the user's surrounding environment, always on, and able to attract the user's attention even when not actively in use. Mann [8] cites three desirable properties for wearabl... | [
1017,
2166
] | Train |
2,087 | 1 | Event-Learning And Robust Policy Heuristics . In this paper we introduce a novel form of reinforcement learning called event-learning or E-learning. Events are ordered pairs of consecutive states. We define the corresponding event-value function. Learning rules which are guaranteed to converge to the optimal event-value function are derived. Combining our method with a known robust control method, the SDS algorithm, we introduce Robust Policy Heuristics (RPH). It is shown that RPH (a fast-adapting non-Markovian policy) is particularly useful for coarse models of the environment and for partially observed systems. Fast adaptation may allow to separate the time scale of learning to control a Markovian process and the time scale of adaptation of a non-Markovian policy. In our E-learning framework the de nition of modules is straightforward. E-learning is well suited for policy switching and planning, whereas RPH alleviates the `curse of dimensionality' problem. Computer simulations of a two-link pendulum with coarse discretization and noisy controller are shown to demonstrate the underlying principle. Date: First version: May 14, 2001, second version: May 19, 2001. Key words and phrases. reinforcement learning, robust control, event representation, continuous dynamical systems, non-Markovian policy. This work was supported by the Hungarian National Science Foundation (Grant OTKA 32487.) Thanks are due to Tor M. Aamodt for providing his double pendulum software [1] and to Gabor Szirtes for careful reading of the manuscript. THIS WORK IS A SHORTENED VERSION OF THE THESIS WORK OF I. POLIK AND I. SZITA SUPERVISED BY A. LORINCZ (SUBMITTED ON DECEMBER 6, 2000) THAT WON FIRST PRIZE IN THE STUDENT COMPETITION IN HUNGARY IN SECTION `MATHEMATICAL METHODS FOR COMPUTERS' ON APRIL 11, 2001 | [
197,
1964
] | Test |
2,088 | 2 | Techniques For Automatic Digital Video Composition Recent developments in digital technology have enabled a class of video-based applications that were not previously viable. However, digital video production systems face the challenge of accessing the inherently linear and time-dependent media of audio and video, and providing effective means of composing them into a cohesive piece for presentation. Moreover, there are no appropriate metrics that allow for assessment of the quality of an automatically-composed video piece. Techniques presently available are limited in scope, and do not account for all the features of a composition. This dissertation presents metrics that evaluate the quality of a video composition. In addition, it proposes techniques for automatic composition of video presentations as well as improvements in access to digital video data. Yet another challenge faced by video production systems is the customization of the presentation to suit user profiles. For instance, certain elements of video compositions, such as v... | [
3031
] | Validation |
2,089 | 1 | The Evolution of the Soar Cognitive Architecture The origins of the Soar architecture can be traced back to the seminal research of Allen Newell and Herbert Simon on symbol systems, heuristic search, goals, problem spaces, and production systems. Since its official inception in 1982, Soar has evolved through six major releases, as both an AI architecture and as the basis for a unified theory of cognition. This paper traces this evolutionary path, starting with Soar's intellectual roots, and then proceeding through the stages defined by the six major system releases. Each stage is characterized with respect to a hierarchy of four levels of analysis: the knowledge level, the problem space level, the symbolic architecture level, and the implementation level. | [] | Train |
2,090 | 3 | Distance Browsing in Spatial Databases Two different techniques of browsing through a collection of spatial objects stored in an R-tree spatial data structure on the basis of their distances from an arbitrary spatial query object are compared. The conventional approach is one that makes use of a k-nearest neighbor algorithm where k is known prior to the invocation of the algorithm. Thus if m#kneighbors are needed, the k-nearest neighbor algorithm needs to be reinvoked for m neighbors, thereby possibly performing some redundant computations. The second approach is incremental in the sense that having obtained the k nearest neighbors, the k +1 st neighbor can be obtained without having to calculate the k +1nearest neighbors from scratch. The incremental approach finds use when processing complex queries where one of the conditions involves spatial proximity (e.g., the nearest city to Chicago with population greater than a million), in which case a query engine can make use of a pipelined strategy. A general incremental nearest neighbor algorithm is presented that is applicable to a large class of hierarchical spatial data structures. This algorithm is adapted to the R-tree and its performance is compared to an existing k-nearest neighbor algorithm for R-trees [45]. Experiments show that the incremental nearest neighbor algorithm significantly outperforms the k-nearest neighbor algorithm for distance browsing queries in a spatial database that uses the R-tree as a spatial index. Moreover, the incremental nearest neighbor algorithm also usually outperforms the k-nearest neighbor algorithm when applied to the k-nearest neighbor problem for the R-tree, although the improvement is not nearly as large as for distance browsing queries. In fact, we prove informally that, at any step in its execution, the incremental... | [
1319
] | Validation |
2,091 | 2 | A Query Paradigm to Discover the Relation between Text and Images This paper studies the relation between images and text in image databases. An analysis of this relation results in the definition of three distinct query modalities: (1) linguistic scenario: images are part of a whole including a self-contained linguistic discourse, and their meaning derives from their interaction with the linguistic discourse. A typical case of this scenario is constituted by images on the World Wide Web; (2) closed world scenario: images are defined in a limited domain, and their meaning is anchored by conventions and norms in that domain. (3) user scenario: the linguistic discourse is provided by the user. This is the case of highly interactive systems with relevance feedback. This paper deals with image databases of the first type. It shows how the relation between images (or parts of images) and text can be inferred, and exploited for search. The paper develops a similarity model in which the similarity between two images is given by both their visual similarity... | [
2283,
2459
] | Train |
2,092 | 0 | Compensatory Negotiation for Agent-Based Project Schedule Coordination Practitioners have tried to solve a project schedule coordination problem involving many subcontractors with a centralized approach, but failed to provide a cohesive solution. They have overlooked a principle that general contractors cannot coordinate subcontractors like they do their own forces. The project schedule coordination problem could be solved better by subcontractors in a distributed manner. This paper presents a formalized negotiation methodology for distributed project schedule coordination --- a framework wherein a project can be rescheduled dynamically by all of the concerned project participants. The compensatory negotiation methodology is developed to allow agents to transfer utility to other agents for compensation of disadvantageous agreements through a multi-linked negotiation process. By employing software agents that are capable of compensatory negotiation, practitioners now can solve the problem and explore and exploit new opportunities an agent-based framework offers. 1. | [] | Train |
2,093 | 4 | Dialogue Management for Multimodal User Registration User registration refers to associating certain personal information with a user. It is widely used in hospitals, hotels and conferences. In this paper, we propose an approach to interactive user registration by combining face recognition, speech recognition and speech synthesis technologies together through an efficient dialogue manager. In order to minimize a user’s effort, we employ a new dialogue management model based on a finite state automaton (FSA), which uses a Baysian network to fuse the user’s information from multiple channels (e.g., face image, speech, records stored in a pre-constructed database) to reliably estimate the confidence about user identity. Instead of fixing weights, the FSA adjusts its weights dynamically by integrating partial information from multiple information sources. This is achieved by maximizing an objective function to determine an optimal action at each succeeding state according to current confidence and information cues. Thus the transition between states can be done along the shortest path from the initial state to the goal state. We have developed a multimodal user registration system to demonstrate the feasibility of the proposed approach. | [] | Train |
2,094 | 0 | Task-Oriented Collaboration with Embodied Agents in Virtual Worlds We are working toward animated agents that can collaborate with human students in virtual worlds. The agent's objective is to help students learn to perform physical, procedural tasks, such as operating and maintaining equipment. Like most of the previous research on task-oriented dialogues, the agent (computer) serves as an expert that can provide guidance to a human novice. Research on such dialogues dates back more than twenty years (Deutsch 1974), and the subject remains an active research area (Allen et al. 1996; Lochbaum 1994; Walker 1996). However, most of that research has focused solely on verbal dialogues, even though the earliest studies clearly showed the ubiquity of nonverbal communication in human task-oriented dialogues (Deutsch 1974). To allow a wider variety of interactions among agents and human students, we use virtual reality (Durlach and Mavor 1995); agents and students cohabit a threedimensional, interactive, simulated mock-up of the student' | [
1492,
1805,
2992
] | Validation |
2,095 | 3 | Improving Min/Max Aggregation over Spatial Objects We examine the problem of computing MIN/MAX aggregate queries over a collection of spatial objects. Each spatial object is associated with a weight (value), for example, the average temperature or rainfall over the area covered by the object. Given a query rectangle, the MIN/MAX problem computes the minimum/maximum weight among all objects intersecting the query rectangle. Traditionally such queries have been performed as range search queries. Assuming that the objects are indexed by a spatial access method, the MIN/MAX is computed as objects are retrieved. This requires effort proportional to the number of objects intersecting the query interval, which may be large. A better approach is to maintain aggregate information among the index nodes of the spatial access method; then various index paths can be eliminated during the range search. In this paper we propose four optimizations that further improve the performance of MIN/MAX queries. Our experiments show that the proposed optimizations offer drastic performance improvement over previous approaches. Moreover, as a by-product of this work we present an optimized version of the MSB-tree, an index that has been proposed for the MIN/MAX computation over 1-dimensional interval objects. | [
817,
1894,
2435,
2806
] | Train |
2,096 | 1 | Content-Based Book Recommending Using Learning for Text Categorization Recommender systems improve access to relevant products and information by making personalized suggestions based on previous examples of a user's likes and dislikes. Most existing recommender systems use collaborative filtering methods that base recommendations on other users' preferences. By contrast, content-based methods use information about an item itself to make suggestions. This approach has the advantage of being able to recommend previously unrated items to users with unique interests and to provide explanations for its recommendations. We describe a content-based book recommending system that utilizes information extraction and a machine-learning algorithm for text categorization. Initial experimental results demonstrate that this approach can produce accurate recommendations. KEYWORDS: Recommender systems, information filtering, machine learning, text categorization INTRODUCTION There is a growing interest in recommender systems that suggest music, films, books, and othe... | [
1478,
1958,
2100,
2300,
2631
] | Test |
2,097 | 2 | Information Visualization within a Digital Video Library . The Informedia Digital Video Library contains over a thousand hours of video, consuming over a terabyte of disk space. This paper summarizes the multimedia abstractions used to represent this video in prior systems and introduces the visualization techniques employed to browse and navigate multiple video documents at once. Keywords: digital video library, information visualization, multimedia abstraction 1. Introduction The Informedia Project at Carnegie Mellon University deals primarily with video. The goal of the project is to enable full content search and retrieval from digital video libraries (Christel et al., 1996; Wactlar et al., 1997). Consider the task of trying to find a five-minute video clip of interest from a library of a thousand hour-long videotapes. In the analog domain, this task would be interminable and the frustrated user would probably walk away without completing the task. Simply digitizing the video will not make the job easier. Through the use of speech rec... | [
1666
] | Train |
2,098 | 1 | Integrating Multiple Classifiers In Visual Object Detectors Learned From User Input There have been many recent efforts in contentbased retrieval to perform automatic classification of images/visual objects. Most approaches, however, have focused on using individual classifiers. In this paper, we study the way in which, in a dynamic framework, multiple classifiers can be combined when applying Visual Object Detectors. We propose a hybrid classifier combination approach, in which decisions of individual classifiers are combined in the following three ways: (1) classifier fusion, (2) classifier cooperation, and (3) hierarchical combination. In earlier work, we presented the Visual Apprentice framework, in which a user defines visual object models via a multiple-level object-definition hierarchy (region, perceptual-area, object part, and object). As the user provides examples from images or videos, visual features are extracted and multiple classifiers are learned for each node of the hierarchy. In this paper, we discuss the benefits of hybrid classifier combination in the Visual Apprentice framework, and show some experimental results in classifier fusion. These results suggest possible improvements in classification accuracy, particularly of detectors reported earlier for Baseball video, images with skies, and images with handshakes. | [
3031
] | Train |
2,099 | 0 | The Security Architecture of the M&M Mobile Agent Framework In the Mobile Agent programming model, small threads of execution migrate from machine to machine, performing their operations locally. For being able to deploy such a model into real world applications, security is a vital concern. In the M&M project we have developed a system that departures from the traditional platform-based execution model for mobile agents. In M&M there are no agent platforms. Instead there is a component framework that allows the applications to become able of sending and receiving agents by themselves in a straightforward manner. In this paper we examine the security mechanisms available in M&M, and how integration with existing applications is done. One difficult aspect of this work is that all the features must work with the security mechanisms that already exist on the applications. This is so because the components are integrated from within into the applications, which already have security mechanisms in place. Currently, M&M provides features like fine-grain security permissions, encryption of agents and data, certificate distribution using LDAP and cryptographic primitives for agents. For validating the approach and solutions found, we have integrated the framework into several off-the-shelf web servers, having the security mechanisms running, with no problems. | [
589,
2128,
2224
] | Validation |
2,100 | 1 | Learning to Classify Text from Labeled and Unlabeled Documents In many important text classification problems, acquiring class labels for training documents is costly, while gathering large quantities of unlabeled data is cheap. This paper shows that the accuracy of text classifiers trained with a small number of labeled documents can be improved by augmenting this small training set with a large pool of unlabeled documents. We present a theoretical argument showing that, under common assumptions, unlabeled data contain information about the target function. We then introduce an algorithm for learning from labeled and unlabeled text based on the combination of Expectation-Maximization with a naive Bayes classifier. The algorithm first trains a classifier using the available labeled documents, and probabilistically labels the unlabeled documents; it then trains a new classifier using the labels for all the documents, and iterates to convergence. Experimental results, obtained using text from three different realworld tasks, show that the use of unlabeled data reduces classification error by up to 33%. | [
759,
809,
888,
1478,
1601,
1821,
1823,
2033,
2096,
2461,
2676,
2808,
2918,
2947,
3115
] | Validation |
2,101 | 1 | Comparing Structures using a Hopfield-style Neural Network Labeled graphs are an appropriate and popular representation of structured objects in many domains. If the labels describe the properties of real world objects and their relations, finding the best match between two graphs turns out to be the weakly defined, NP-complete task of establishing a mapping between them that maps similar parts onto each other preserving as much as possible of their overall structural correspondence. In this paper, former approaches of structural matching and constraint relaxation by spreading activation in neural networks and the method of solving optimization tasks using Hopfield-style nets are combined. The approximate matching task is reformulated as the minimization of a quadratic energy function. The design of the approach enables the user to change the parameters and the dynamics of the net so that knowledge about matching preferences is included easily and transparently. In the last section, some examples demonstrate the successful application of ... | [
2048
] | Train |
2,102 | 4 | Biologically Motivated Distributed Design for Adaptive Knowledge Management We discuss how distributed designs that draw from biological network metaphors can largely improve the current state of information retrieval and knowledge management of distributed information systems. In particular, two adaptive recommendation systems named TalkMine and @ApWeb are discussed in more detail. TalkMine operates at the semantic level of keywords. It leads different databases to learn new and adapt existing keywords to the categories recognized by its communities of users using distributed algorithms. | [
471,
1071
] | Train |
2,103 | 1 | Non-Standard Crossover for a Standard Representation -- Commonality-Based Feature Subset Selection The Commonality-Based Crossover Framework has been presented as a general model for designing problem specific operators. Following this model, the Common Features/Random Sample Climbing operator has been developed for feature subset selection--a binary string optimization problem. Although this problem should be an ideal application for genetic algorithms with standard crossover operators, experiments show that the new operator can find better feature subsets for classifier training. 1 INTRODUCTION A classification system is used to predict the decision class of an object based on its features. When training a classifier, it is beneficial to use only the features relevant to prediction accuracy, and to ignore the irrelevant features [Koh95]. The benefit arises from an increase in the "signalto -noise ratio" of the data, and a reduction in the time required to train the classifier. Thus, the objective of feature subset selection is to identify the (most) relevant features. Feature sub... | [
137
] | Train |
2,104 | 2 | Representing Sentence Structure in Hidden Markov Models for Information Extraction We study the application of Hidden Markov Models (HMMs) to learning information extractors for ¤-ary relations from free text. We propose an approach to representing the grammatical structure of sentences in the states of the model. We also investigate using an objective function during HMM training which maximizes the ability of the learned models to identify the phrases of interest. We evaluate our methods by deriving extractors for two binary relations in biomedical domains. Our experiments indicate that our approach learns more accurate models than several baseline approaches. 1 | [
603,
2443,
2522
] | Validation |
2,105 | 3 | The 3W Model and Algebra for Unified Data Mining Real data mining/analysis applications call for a framework which adequately supports knowledge discovery as a multi-step process, where the input of one mining operation can be the output of another. Previous studies, primarily focusing on fast computation of one specific mining task at a time, ignore this vital issue. Motivated by this observation, we develop a unified model supporting all major mining and analysis tasks. Our model consists of three distinct worlds, corresponding to intensional and extensional dimensions, and to data sets. The notion of dimension is a centerpiece of the model. Equipped with hierarchies, dimensions integrate the output of seemingly dissimilar mining and analysis operations in a clean manner. We propose an algebra, called the dimension algebra, for manipulating (intensional) dimensions, as well as operators that serve as "bridges" between the worlds. We demonstrate by examples that several real data mining processes can be captured ... | [
1920
] | Train |
2,106 | 2 | Web Interaction and the Navigation Problem in Hypertext written for Encyclopedia of Microcomputers The web has become a ubiquitous tool, used in day-to-day work, to find information and conduct business, and it is revolutionising the role and availability of information. One of the problems encountered in web interaction, which is still unsolved, is the navigation problem, whereby users can "get lost in hyperspace", meaning that when following a sequence of links, i.e. a trail of information, users tend to become disoriented in terms of the goal of their original query and the relevance to the query of the information they are currently browsing. Herein we build statistical foundations for tackling the navigation problem based on a formal model of the web in terms of a probabilistic automaton, which can also be viewed as a finite ergodic Markov chain. In our model of the web the probabilities attached to state transitions have two interpretations, namely, they can denote the proportion of times a user followed a link, and alternatively they can denote the expected utility of following a link. Using this approach we have developed two techniques for constructing a web view based on the two interpretations of the probabilities of links, where a web view is a collection of relevant trails. The first method we describe is concerned with finding frequent user behaviour patterns. A collection of trails is taken as input and an ergodic Markov chain is produced as output with the probabilities of transitions corresponding to the frequency the user traversed the associated links. The second method we describe is a reinforcement learning algorithm that attaches higher probabilities to links whose expected trail relevance is higher. The user's home page and a query are taken as input and an ergodic Markov chain is produced as output with the probabilities of... | [
471,
1201,
1431,
1930,
2503,
2545,
2984
] | Train |
2,107 | 1 | A Bayesian Computer Vision System for Modeling Human Interactions AbstractÐWe describe a real-time computer vision and machine learning system for modeling and recognizing human behaviors in a visual surveillance task [1]. The system is particularly concerned with detecting when interactions between people occur and classifying the type of interaction. Examples of interesting interaction behaviors include following another person, altering one's path to meet another, and so forth. Our system combines top-down with bottom-up information in a closed feedback loop, with both components employing a statistical Bayesian approach [2]. We propose and compare two different state-based learning architectures, namely, HMMs and CHMMs for modeling behaviors and interactions. The CHMM model is shown to work much more efficiently and accurately. Finally, to deal with the problem of limited training data, a synthetic ªAlife-styleº training system is used to develop flexible prior models for recognizing human interactions. We demonstrate the ability to use these a priori models to accurately classify real human behaviors and interactions with no additional tuning or training. Index TermsÐVisual surveillance, people detection, tracking, human behavior recognition, Hidden Markov Models. 1 | [
51,
690,
919,
2316,
2906
] | Validation |
2,108 | 2 | User Modeling for Information Access Based on Implicit Feedback User modeling can be used in information filtering and retrieval systems to improve the representation of a user's information needs. User models can be constructed by hand, or learned automatically based on feedback provided by the user about the relevance of documents that they have examined. By observing user behavior, it is possible to infer implicit feedback without requiring explicit relevance judgments. Previous studies based on Internet discussion groups (USENET news) have shown reading time to be a useful source of implicit feedback for predicting a user's preferences. The study reported in this paper extends that work by providing framework for considering alternative sources of implicit feedback, examining whether reading time is useful for predicting a user's preferences for academic and professional journal articles, and exploring whether retention behavior can usefully augment the information that reading time provides. Two user studies were conducted in which undergradua... | [
2503
] | Train |
2,109 | 2 | Language Models for Financial News Recommendation We present a unique approach to identifying news stories that influence the behavior of financial markets. Specifically, we describe the design and implementation of Ænalyst, a system that can recommend interesting news stories -- stories that are likely to affect market behavior. Ænalyst operates by correlating the content of news stories with trends in financial time series. We identify trends in time series using piecewise linear fitting and then assign labels to the trends according to an automated binning procedure. We use language models to represent patterns of language that are highly associated with particular labeled trends. &Aelig;nalyst can then identify and recommend news stories that are highly indicative of future trends. We evaluate the system in terms of its ability to recommend the stories that will a ect the behavior of the stock market. We demonstrate that stories recommended by &Aelig;nalyst could be used to profitably predict forthcoming trends in stock prices. | [
1544,
2255,
3107
] | Validation |
2,110 | 3 | Guaranteeing No Interaction between Functional Dependencies and Tree-Like Inclusion Dependencies Functional dependencies (FDs) and inclusion dependencies (INDs) are the most fundamental integrity constraints that arise in practice in relational databases. A given set of FDs does not interact with a given set of INDs if logical implication of any FD can be determined solely by the given set of FDs, and logical implication of any IND can be determined solely by the given set of INDs. The set of tree-like INDs constitutes a useful subclass of INDs whose implication problem is polynomial time decidable. We exhibit a necessary and sufficient condition for a set of FDs and tree-like INDs not to interact; this condition can be tested in polynomial time. 1 Introduction The implication problem for FDs and INDs is the problem of deciding for a given set \Sigma of FDs and INDs whether \Sigma logically implies oe, where oe is an FD or an IND. The implication problem is central in data dependency theory and is also utilised in the process of database design, since it can be used to test wheth... | [
1942
] | Validation |
2,111 | 1 | 1BC: a First-Order Bayesian Classifier . In this paper we present 1BC, a first-order Bayesian Classifier. Our approach is to view individuals as structured terms, and to distinguish between structural predicates referring to subterms (e.g. atoms from molecules), and properties applying to one or several of these subterms (e.g. a bond between two atoms). We describe an individual in terms of elementary features consisting of zero or more structural predicates and one property; these features are considered conditionally independent following the usual naive Bayes assumption. 1BC has been implemented in the context of the first-order descriptive learner Tertius, and we describe several experiments demonstrating the viability of our approach. 1 Introduction In this paper we present 1BC, a first-order Bayesian Classifier. While the propositional Bayesian Classifier makes the naive Bayes assumption of statistical independence of elementary features (one attribute taking on a particular value) given the class value, it is not i... | [
2140,
2467
] | Test |
2,112 | 3 | Improving Cross Language Retrieval with Triangulated Translation Most approaches to cross language information retrieval assume that resources providing a direct translation between the query and document languages exist. This paper presents research examining the situation where such an assumption is false. Here, an intermediate (or pivot) language provides a means of transitive translation of the query language to that of the document via the pivot, at the cost, however, of introducing much error. The paper reports the novel approach of translating in parallel across multiple intermediate languages and fusing the results. Such a technique removes the error, raising the effectiveness of the tested retrieval system, up to and possibly above the level expected, had a direct translation route existed. Across a number of retrieval situations and combinations of languages, the approach proves to be highly effective. | [
2186
] | Train |
2,113 | 4 | Wearing It Out: First Steps Toward Mobile Augmented Reality Systems Introduction Over the past decade, there has been a ground swell of activityintwo #elds of user interface research: augmented reality and wearable computing. Augmentedreality #1# refers to the creation of virtual environments that supplement, rather than replace, the real world with additional information. This is accomplished through the use of #see-through" displays that enrich the user's view of the world byoverlaying visual, auditory, and even haptic, material on what she experiences. Visual augmented reality systems typically, but not exclusively, employ head-tracked, head-worn displays. These either use half-silvered mirror beam splitters to re#ect small computer displays, optically combining them with a view of the real world, or use opaque displays fed by electronics that merge imagery captured by head-worn cameras with synthesized graphics. Wearable computing moves computers o# the desktop and onto the user's body, made possible through the | [
1740
] | Validation |
2,114 | 4 | Gestural Interface to a Visual Computing Environment for Molecular Biologists In recent years there has been tremendous progress in 3D, immersive display and virtual reality (VR) technologies. Scientific visualization of data is one of many applications that has benefited from this progress. To fully exploit the potential of these applications in the new environment there is a need for "natural" interfaces that allow the manipulation of such displays without burdensome attachments. This paper describes the use of visual hand gesture analysis enhanced with speech recognition for developing a bimodal gesture/speech interface for controlling a 3-D display. The interface augments an existing application, VMD, which is a VR visual computing environment for molecular biologists. The free hand gestures are used for manipulating the 3-D graphical display together with a set of speech commands. We concentrate on the visual gesture analysis techniques used in developing this interface. The dual modality of gesture/speech is found to greatly aid the interaction capability.... | [
1454
] | Test |
2,115 | 0 | Secure Mobile Agent using Strong Non-designated Proxy Signature It is expected that mobile agent will be widely used for electronic commerce as an important key technology. If a mobile agent can sign a message in a remote server on behalf of a customer without exposing his/her private key, it can be used not only to search for special products or services, but also to make a contract with a remote server. To construct mobile agents, [KBC00] used an RSA-based undetachable signature scheme, but it does not provide server's non-repudiation because the undetachable signature does not contain server's signature. Mobile agent is a very good application example of proxy signature, and the undetachable signature can be considered as an example of proxy signature. In this paper we show that secure mobile agent can be constructed using strong non-designated proxy signature [LKK01] which represents both the original signer's (customer) and the proxy signer's (remote server) signatures. We provide RSA-based and Schnorr-based constructions of secure mobile agent, and moreover we show that the Schnorr-based scheme can be used very eciently in multi-proxy mobile agent situation. Keywords. Secure mobile agent, strong non-designated proxy signature, multi-proxy signature. 1 | [
2969
] | Train |
2,116 | 2 | Half-pipe Anchoring: An Efficient Technique for Multiple Connection Handoff This paper presents Half--pipe anchoring, a novel technique to build a multiple connection handoff mechanism that enables efficient use of resources in a server cluster, improves the scalability of the cluster and supports construction of heterogeneous cluster architectures where nodes are specialized to perform specific tasks of client requests efficiently. The key idea behind our approach is to decouple the two unidirectional half-pipes that make up a TCP connection between a client and a server and anchor the unidirectional half--pipe from the client to the cluster at a designated server while allowing the half--pipe from the cluster to the client to migrate on a per--request basis to an optimal server where the request is best serviced. We describe the design and implementation of a prototype multiple connection handoff mechanism in the Linux kernel and demonstrate the benefits of our technique. | [
1179
] | Validation |
2,117 | 1 | Feature Weighting for Lazy Learning Algorithms : Learning algorithms differ in the degree to which they process their inputs prior to their use in performance tasks. Many algorithms eagerly compile input samples and use only the compilations to make decisions. Others are lazy: they perform less precompilation and use the input samples to guide decision making. The performance of many lazy learners significantly degrades when samples are defined by features containing little or misleading information. Distinguishing feature relevance is a critical issue for these algorithms, and many solutions have been developed that assign weights to features. This chapter introduces a categorization framework for feature weighting approaches used in lazy similarity learners and briefly surveys some examples in each category. 1.1 INTRODUCTION Lazy learning algorithms are machine learning algorithms (Mitchell, 1997) that are welcome members of procrastinators anonymous. Purely lazy learners typically display the following characteristics (Aha, 19... | [
1737,
2641,
2823
] | Validation |
2,118 | 4 | Supporting Group Collaboration with Inter-Personal Awareness Devices . An Inter-Personal Awareness Device, or IPAD, is a hand-held or wearable device designed to support awareness and collaboration between people who are in the physical vicinity of each other. An IPAD is designed to supply constant awareness information to users in any location without relying on an underlying infrastructure. We have constructed one such device, the Hummingbird, which gives members of a group continuous aural and visual indication when other group members are close. We have used the Hummingbirds in several different situations to explore how they affect group awareness. These experiences indicated that the Hummingbird increased awareness between group members, and that it could complement other forms of communication such as phone and e-mail. In particular, we found the Hummingbird to be useful when a group of people were in an unfamiliar location, for instance during a trip, where no other communication support was available. We argue that IPADs such as th... | [
457,
1156,
1634,
1994,
2227
] | Test |
2,119 | 4 | Situation Aware Computing with Wearable Computers 1 Motivation for contextual aware computing: For most computer systems, even virtual reality systems, sensing techniques are a means of getting input directly from the user. However, wearable sensors and computers offer a unique opportunity to re-direct sensing technology towards recovering more general user context. Wearable computers have the potential to "see" as the user sees, "hear" as the user hears, and experience the life of the user in a "first-person" sense. This increase in contextual and user information may lead to more intelligent and fluid interfaces that use the physical world as part of the interface. Wearable computers are excellent platforms for contextually aware applications, but these applications are also necessary to use wearables to their fullest. Wearables are more than just highly portable computers, they perform useful work even while the wearer isn't directly interacting with the system. In such environments the user needs to concentrate on his environment, not on the computer interface, so the wearable needs to use information from the wearer's context to be the least distracting. For example, imagine an interface which is aware of the user's location: while being in the subway, the system might alert him with a | [
727
] | Train |
2,120 | 3 | Active Views for Electronic Commerce Electronic commerce is emerging as a major Web-supported application. In this paper we argue that database technology can, and should, provide the backbone for a wide range of such applications. More precisely, we present here the ActiveViews system, which, relaying on an extensive use of database features including views, active rules (triggers), and enhanced mechanisms for notification, access control and logging/tracing of users activities, provides the needed basis for electronic commerce. Based on the emerging XML standards (DOM, query languages for XML, etc.), the system offers a novel declarative view specification language, describing the relevant data and activities of all actors (e.g. vendors and clients) participating in electronic commerce activities . Then, acting as an application generator, the system generates an actual, possibly customized, Web application that allows users to perform the given set of controlled activities and to work interactively on the specified dat... | [
1575,
1667,
2329
] | Train |
2,121 | 3 | Temporal Patterns (TRAPs) in ASR of Noisy Speech In this paper we study a new approach to processing temporal information for automatic speech recognition (ASR). Specifically, we study the use of rather longtime TempoRAl Patterns (TRAPs) of spectral energies in place of the conventional spectral patterns for ASR. The proposed Neural TRAPs are found to yield significant amount of complementary information to that of the conventional spectral feature based ASR system. A combination of these two ASR systems is shown to result in improved robustness to several types of additive and convolutive environmental degradations. 1. INTRODUCTION 1.1. Spectral features Spectrum-based techniques form the basis of most feature extraction methods in current ASR. A drawback of the spectral features is that they are quite sensitive to changes in the communication environment e.g. characteristics of different communication channels or environmental noise. Subsequently, recognizers based on spectral features exhibit rapid degradation in performance in ... | [
2130
] | Validation |
2,122 | 5 | Remote Agent: To Boldly Go Where No AI System Has Gone Before Renewed motives for space exploration have inspired NASA to work toward the goal of establishing a virtual presence in space, through heterogeneous effets of robotic explorers. Information technology, and Artificial Intelligence in particular, will play a central role in this endeavor by endowing these explorers with a form of computational intelligence that we call remote agents. In this paper we describe the Remote Agent, a specific autonomous agent architecture based on the principles of model-based programming, on-board deduction and search, and goal-directed closed-loop commanding, that takes a significant step toward enabling this future. This architecture addresses the unique characteristics of the spacecraft domain that require highly reliable autonomous operations over long periods of time with tight deadlines, resource constraints, and concurrent activity among tightly coupled subsystems. The Remote Agent integrates constraint-based temporal planning and scheduling, robust multi-threaded execution, and model-based mode identification and reconfiguration. The demonstration of the integrated system as an on-board controller for Deep Space One, NASA's rst New Millennium mission, is scheduled for a period of a week in late 1998. The development of the Remote Agent also provided the opportunity to reassess some of AI's conventional wisdom about the challenges of implementing embedded systems, tractable reasoning, and knowledge representation. We discuss these issues, and our often contrary experiences, throughout the paper. | [
111,
1668,
3154
] | Train |
2,123 | 2 | Information Extraction: Beyond Document Retrieval In this paper we give a synoptic view of the growth text processing technology of information extraction (IE) whose function is to extract information about a pre-specified set of entities, relations or events from natural language texts and to record this information in structured representations called templates. Here we describe the nature of the IE task, review the history of the area from its origins in AI work in the 1960's and 70's till the present, discuss the techniques being used to carry out the task, describe application areas where IE systems are or are about to be at work, and conclude with a discussion of the challenges facing the area. What emerges is a picture of an exciting new text processing technology with a host of new applications, both on its own and in conjunction with other technologies, such as information retrieval, machine translation and data mining. | [
1295
] | Train |
2,124 | 1 | Learning Case Classification for Improving Case-Based Reasoning this paper, we consider unconstrained graphs which may be either directed or undirected and, possibly, labelled or unlabelled. Computing structural similarity is essentially based on graph matching which actually means computing subgraph isomorphism. This subproblem is known to be NP-complete. It becomes more tractable when using labelled and directed graphs. The particular approach developed and investigated throughout the present workshop presentation can be easily generalized to other structural similarity concepts. For readability, we confine ourselves to graph representations. Because of NP-completeness of the subgraph isomorphism problem, classical case retrieval is extremely expensive, if every member of a usually huge case base is potentially queried. The performance of this search can be increased enormously by prestructuring the case base in case classes, i.e. clustering. Preferably, a given case base CB is clustered with respect to structural similarity, i.e. graphs of a close structural relationship are grouped together and represented by the graph(s) describing their common structural similarity. Let us assume that any given case base CB is separated in n partitions or classes. Each class CB i (i = 1; :::; n) consists of a set of graph-represented cases and is determined by a graph \Theta | [
1844
] | Train |
2,125 | 1 | Multilevel Simulation and Numerical Optimization of Complex Engineering Designs Multilevel representations have been studied extensively by artificial intelligence researchers. We present a general method that utilizes the multilevel paradigm to attack the problem of performing multidiscipline engineering design optimization in the presence of many local optima. The method uses a multidisciplinary simulator at multiple levels of abstraction, paired with a multilevel search space. We tested the method in the domain of conceptual design of supersonic transport aircraft, focusing on the airframe and the exhaust nozzle, and using sequential quadratic programming as the optimizer at each level. We found that using multilevel simulation and optimization can decrease the cost of design space search by an order of magnitude. 1 Introduction A major barrier to the use of gradient-based search methods for engineering design is that complex, multidisciplinary design spaces tend An earlier version of this article was presented at 6 th AIAA/NASA/USAF Multidisciplinary Ana... | [
144,
1886
] | Train |
2,126 | 1 | Constructive Neural Network Learning Algorithms for Multi-Category Pattern Classification Constructive learning algorithms offer an attractive approach for the incremental construction of near-minimal neural-network architectures for pattern classification. They help overcome the need for ad hoc and often inappropriate choices of network topology in algorithms that search for suitable weights in a priori fixed network architectures. Several such algorithms are proposed in the literature and shown to converge to zero classification errors (under certain assumptions) on tasks that involve learning a binary to binary mapping (i.e., classification problems involving binary-valued input attributes and two output categories). We present two constructive learning algorithms MPyramid-real and MTiling-real that extend the pyramid and tiling algorithms, respectively, for learning real to M-ary mappings (i.e., classification problems involving real-valued input attributes and multiple output classes). We prove the convergence of these algorithms and empirically demonstrate their applicability to practical pattern classification problems. Additionally, we show how the incorporation of a local pruning step can eliminate several redundant neurons from MTiling-real networks. | [
354,
428,
1915,
2046
] | Test |
2,127 | 4 | Plans as Situated Action: An Activity Theory Approach to Workflow Systems : Within the community of CSCW the notion and nature of workflow systems as prescriptions of human work has been debated and criticised. Based on the work of Suchman (1987) the notion of situated action has often been viewed as opposed to planning work. Plans, however, do play an essential role in realising work. Based on experiences from designing a computer system that supports the collaboration within a hospital, this paper discusses how plans themselves are made out of situated action, and in return are realised in situ. Thus, work can be characterised as situated planning. This understanding is backed up by Activity Theory, which emphasises the connection between plans and the contextual conditions for realising these plans in actual work. Introduction The issue of workflow systems has been addressed by several authors as ways of routing information objects among users, and to specify automatic actions to be taken in that routing typically according to certain process models (Me... | [
1695,
2332
] | Train |
2,128 | 0 | Providing Applications with Mobile Agent Technology Over the last couple of years we have been working on the development of mobile agents systems and its application to the areas of telecommunications and network management. This work path produced positive results: a competitive mobile agent platform was built, the run-time benefits of mobile agents were proved, and our industrial partners have developed practical applications that are being integrated into commercial products. However, | [
589,
1539,
2099,
2709
] | Validation |
2,129 | 0 | A Language for Exchanging Agent UML Protocol Diagrams For several years, interaction protocol designers have a new formalism which takes into account multiagent system features: autonomy, cooperation, etc. This formalism is called Agent UML [6]. For the moment, designers can describe protocols with the Agent UML protocol diagrams but they do not have a textual language in order to exchange protocols or to check properties on them. The aim of this paper is to provide such a language. This language is called AXF (Agent UML eXchange Format) and is structured as an XML file. This paper presents the syntax of this language and applies AXF to the example of the English Auction Protocol. This paper is published as the technical report ULCS-02-009 from the department of computer science, University of Liverpool. | [
567,
1409,
2944
] | Train |
2,130 | 3 | Traps - Classifiers Of Temporal Patterns The work proposes a radically different set of features for ASR where TempoRAl Patterns of spectral energies are used in place of the conventional spectral patterns. The approach has several inherent advantages, among them robustness to stationary or slowly varying disturbances. 1. INTRODUCTION 1.1. Spectral features In 1665 Isaac Newton made the following observation: 'The filling of a very deepe flaggon with a constant streame of beere or water sounds yer vowells in this order w, u, !, o, a, e, i, y' [8]. What young Newton observed was the spectral resonance peak which enhanced the spectrum of the beer pouring sound and moved up in frequency as the "deepe flaggon" was filling up. Since then, attempts to find acoustic correlates of phonetic categories mostly followed Newton's lead and studied the spectrum of speech. Spectrum-based techniques form the basis of most feature extraction methods in current ASR. A problem with the spectrum of sound is that it can easily be modified by v... | [
2121
] | Train |
2,131 | 3 | Temporal FDs on Complex Objects ing with credit is permitted. To copy otherwise, to republish, to Post on servers, or to redistribute to lists, requires prior speci#c permission and#or a fee. Request permissions from Publications Dept, ACM Inc., fax +1 #212# 869-0481, or permissions@acm.org. 2 # J. Wijsen in reality, regardless of the past and future. In many applications, information about the past and future is just as important as information about the present. Incorporating the time dimension into existing database theory and practice is interesting and important. Temporal databases extend classical #snapshot" databases by supporting the storage and the access of time-related information. When history is taken into account, integrity constraints can place restrictions on the evolution of data in time. This paper deals with extending dependency theory for temporal databases with complex objects. In general, classical dependencies can be applied to temporal databases in a straightforward manner. From an abstr... | [
1072,
2485,
3105
] | Test |
2,132 | 1 | Invariant Fourier-Wavelet Descriptor For Pattern Recognition We present a novel set of descriptors for recognizing complex patterns such as roadsigns, keys, aircrafts, characters, etc. Given a pattern, we first transform it to polar coordinate (r; `) using the centre of mass of the pattern as origin. We then apply the Fourier transform along the axis of polar angle ` and the wavelet transform along the axis of radius r. The features thus obtained are invariant to translation, rotation, and scaling. As an example, we apply the method to a database of 85 printed Chinese characters. The result shows that the Fourier-Wavelet descriptor is an efficient representation which can provide for reliable recognition. Feature Extraction, Fourier Transform, Invariant Descriptor, Multiresolution Analysis, Pattern Recognition, Wavelet Transform. 1 Introduction Feature extraction is a crucial processing step for pattern recognition (15) . Some authors (5\Gamma7;13) extract 1-D features from 2-D patterns. The advantage of this approach is that we can save spa... | [
1046
] | Train |
2,133 | 2 | Clipping and Analyzing News Using Machine Learning Techniques Generating press clippings for companies manually requires a considerable amount of resources. We describe a system that monitors online newspapers and discussion boards automatically. The system extracts, classifies and analyzes messages and generates press clippings automatically, taking the specific needs of client companies into account. Key components of the system are a spider, an information extraction engine, a text classifier based on the Support Vector Machine that categorizes messages by subject, and a second classifier that analyzes which emotional state the author of a newsgroup posting was likely to be in. By analyzing large amount of messages, the system can summarize the main issues that are being reported on for given business sectors, and can summarize the emotional attitude of customers and shareholders towards companies. | [
439,
603,
815,
2342,
2522,
2765
] | Validation |
2,134 | 5 | Cooperative Multiagent Robotic Systems Introduction Teams of robotic systems at first glance might appear to be more trouble than they are worth. Why not simply build one robot that is capable of doing everything we need? There are several reasons why two robots (or more) can be better than one: ffl Distributed Action: Many robots can be in many places at the same time ffl Inherent Parallelism: Many robots can do many, perhaps different things at the same time ffl Divide and Conquer: Certain problems are well suited for decomposition and allocation among many robots ffl Simpler is better: Often each agent in a team of robots can be simpler than a more comprehensive single robot solution No doubt there are more reasons as well. Unfortunately there are also drawbacks, in particular regarding coordination and elimination of interference. The degree of difficulty imposed depends heavily upon the task and the communication and control strategies chosen. | [
963
] | Train |
2,135 | 2 | Distance Metrics and Indexing Strategies for a Digital Library of Popular Music Introduction Digital libraries until now could hardly be described as popular: they tend to be based on esoteric, scholarly sources close to the interests of digital library researchers themselves. We are developing a digital library containing the quintessence of popular culture: music. The principal mode of searching this library will be by sung query: the system should be able to answer the kinds of queries that shop assistants in music stores deal with every day, where a customer can sing a tune, but can't remember the title or artist. The operation of our musical digital library is sketched in Figure 1. At the time of collection creation, MIDI files are gathered from the internet, and indexed based on their note sequences. At retrieval time, a user's sung query is transformed from a waveform to a sequence of pitch-duration events by a pitch tracker. We are using an off-the-shelf pitch tracker: pitch tracking is beyond the scope of the current project. Pot | [
42
] | Validation |
2,136 | 5 | A Java/CORBA based Visual Program Composition Environment for PSEs A Problem Solving Environment (PSE) is a complete, integrated computing environment for composing, compiling and running applications in a specific problem area or domain. Parts of the PSE are domain independent, such as the Visual Programming Composition Environment (VPCE), which may be used for constructing application in a number of different domains, however, other parts are domain specific, such as rules to support particular types of components. A domain independent VPCE is first described, which serves as a user interface for a PSE, and uses Java and CORBA to provide a framework of tools to enable the construction of scientific applications from components. The VPCE consists of a component repository, from which the user can select off-the-shelf or in-house components, a graphical composition area on which components can be combined, various tools that facilitate the configuration of components, the integration of legacy codes into components and the design and bui... | [
1407
] | Validation |
2,137 | 4 | First Steps Towards an Event-Based Infrastructure for Smart Things In this paper, we examine requirements for an infrastructure that supports implementation and deployment of smart things in the real world. We describe a case study (RFID Chef) where kitchen items and ingredients, equipped with remotely accessible electronic tags, drive an interactive context-aware recipe finder through the use of an event-based infrastructure. | [
241,
1035,
2679
] | Train |
2,138 | 0 | Violation Contexts and Deontic Independence . In this paper we discuss the role of context and independence in normative reasoning. First, deontic operators -- obligations, prohibitions, permissions -- referring to the ideal context may conflict with operators referring to a violation (or contrary-to-duty) context. Second, deontic independence is a powerful concept to derive deontic operators from such operators of other violation contexts. These two concepts are used to determine how to proceed once a norm has been violated, a key issue of deontic logic applications in computer science. We also show how violation contexts and deontic independence can be used to give a new analysis of several notorious paradoxes of deontic logic. 1 Introduction Deontic logic is a modal logic in which flp is read as `p ought to be (done),' Fp as `p is forbidden to be (done)' and Pp as `p is permitted to be (done).' Deontic logic has traditionally been used by philosophers to analyze the structure of the normative use of language. In the eightie... | [] | Train |
2,139 | 3 | Query Optimization in the Presence of Limited Access Patterns 1 Introduction The goal of a query optimizer of a database system is to translate a declarative query expressed on a logical schema into an imperative query execution plan that accesses the physical storage of the data, and applies a sequence of relational operators. In building query execution plans, traditional relational query optimizers try to find the most efficient method for accessing the necessary data. When possible, a query optimizer will use auxiliary data structures such as an index on a file in order to efficiently retrieve a certain set of tuples in a relation. However, when such structures do not exist or are not useful for the given query, the alternative of scanning the entire relation always exists. The existence of the fall back option to perform a complete scan is an important assumption in traditional query optimization. Several recent query processing applications have the common characteristic that it is not always possible to perform complete scans on the data. Instead, the query optimization problem is complicated by the fact that there are only limited access patterns to the data. One such | [
923,
930,
1182,
1183,
2022,
2149,
2633,
3062
] | Test |
2,140 | 5 | Strongly Typed Inductive Concept Learning . In this paper we argue that the use of a language with a type system, together with higher-order facilities and functions, provides a suitable basis for knowledge representation in inductive concept learning and, in particular, illuminates the relationship between attribute-value learning and inductive logic programming (ILP). Individuals are represented by closed terms: tuples of constants in the case of attribute-value learning; arbitrarily complex terms in the case of ILP. To illustrate the point, we take some learning tasks from the machine learning and ILP literature and represent them in Escher, a typed, higher-order, functional logic programming language being developed at the University of Bristol. We argue that the use of a type system provides better ways to discard meaningless hypotheses on syntactic grounds and encompasses many ad hoc approaches to declarative bias. 1. Motivation and scope Inductive concept learning consists of finding mappings of individuals (or objects... | [
2111,
2307,
2467
] | Validation |
2,141 | 0 | Revisiting Rationality for Agents With Intentions Formal frameworks for the specification of autonomous agents are commonly based on logics of intention and belief. Desirable properties for logics of intention are particularly non-standard, even more so than for logics of belief. In this paper, we address problems with existing logics of intention and belief by shifting to a non-classical semantics, making use of Rantala's impossible, or non-normal, worlds. Our framework invalidates the problematic properties of intention and, by imposing certain constraints on the algebraic structure of the models, we show that that many desirable properties can be obtained. The non-normal worlds framework provides a fine-grained semantics and proves to be an extremely powerful and flexible tool for the logical specification of rational agent behaviour. 1. Introduction Logics of Belief, Desire and Intention (BDI ) have recently received much attention in the AI literature on the design of autonomous intelligent agents (e.g. [2, 3, 6]). The importanc... | [
2364
] | Validation |
2,142 | 5 | Projective Translations and Affine Stereo Calibration This paper investigates the structure of projective translations - rigid translations expressed as homographies in projective space. A seven parameter representation is proposed, which explicitly represents the geometric entities constraining and defining the translation. A practical algebraic method for estimating these parameters is developed. It provides affine calibration of a stereo rig, determines the translation axis, and allows projective translations to be composed. The practical effectiveness of the calibration is evaluated on synthetic and real image data. 1 Introduction The recovery of structure and motion is a basic problem in machine vision. The difficulties increase when uncalibrated cameras are considered, since in this case metric information is missing. Nevertheless, the rigidity of 3D-motions imposes strong consistency constraints. This allows metric information to be extracted. Mathematically, given a set of points reconstructed in projective space, the effect of ... | [
277,
449
] | Train |
2,143 | 4 | Using Transparent Props For Interaction With The Virtual Table The Virtual Table presents stereoscopic graphics to a user in a workbench-like setting. This paper reports on a user interface and new interaction techniques for the Virtual Table based on transparent props---a tracked hand-held pen and a pad. These props, but in particular the pad, are augmented with 3D graphics from the Virtual Table's display. This configuration creates a very powerful and flexible interface for two-handed interaction that can be applied to other back-projected stereographic displays as well: the pad can serve as a palette for tools and controls as well as a window-like see-through interface, a plane-shaped and throughthe -plane tool, supporting a variety of new interaction techniques. 1. INTRODUCTION While the desktop metaphor is well-understood and represents an effective approach to human-computer interaction for documentoriented 2D tasks, transplanting it to 3D reveals inherent limitations (e.g. [8]). In contrast, interfaces that incorporate true 3D input and ... | [
1584,
2365
] | Train |
2,144 | 0 | Reasoning Agents In Dynamic Domains The paper discusses an architecture for intelligent agents based on the use of A-Prolog - a language of logic programs under the answer set semantics. A-Prolog is used to represent the agent's knowledge about the domain and to formulate the agent's reasoning tasks. We outline how these tasks can be reduced to answering questions about properties of simple logic programs and demonstrate the methodology of constructing these programs. Keywords: Intelligent agents, logic programming and nonmonotonic reasoning. 1 INTRODUCTION This paper is a report on the attempt by the authors to better understand the design of software components of intelligent agents capable of reasoning, planning and acting in a changing environment. The class of such agents includes, but is not limited to, intelligent mobile robots, softbots, immobots, intelligent information systems, expert systems, and decision-making systems. The ability to design intelligent agents (IA) is crucial for such diverse tasks as ... | [
903,
1192,
1568,
1777
] | Validation |
2,145 | 4 | Automatic Construction of Intelligent Diagrammatic Environments Introduction Graphical user interfaces have become an integral part of almost every modern application type and it can be claimed that they are among the driving forces that have made the computer accessible to non-expert users. However, comparing the use of graphics in existent user interfaces with that in non-computer-based work, the inadequacy of standard GUIs for complex visual communication is revealed: Most GUIs are still WIMP interfaces centered around such simple interaction devices like icons, buttons, menus or image maps. On the contrary, in non-computerbased work rich and highly structured graphical notations prevail. There are diagrammatic languages in almost every technical discipline, for example circuit diagrams, architectural floor plans or chemical formulas, and modern software engineering is embracing all kinds of diagrammatic specification methods. Likewise, non-technical fields use their own well-established diagrammatic systems, for example choreography no | [
2246
] | Validation |
2,146 | 0 | Conflicts in a Simple Autonomy-Based Multi-Agent System This paper shows that, in some situations, conflict can deliberately be left in an autonomy-based multi-agent system. This study, supported by experimental results, has two major outcomes. First, it proves that conflict does not necessarily alter the global outcome of the system in a qualitative way. Second, it shows that it is possible to effect the way the global task is achieved by appropriately modifying the environment of the agents. Introduction Our work fits in the framework of Bottom-Up Artificial Intelligence (Brooks 1986), (Brooks 1991) and more particularly, in that of Autonomous Agents (Pfeifer 1995). We are concerned with collective phenomena and their issues and more precisely, the way to carry out solutions that allow an autonomy-based multi-agent system to achieve a global task by virtue of emergence and self-organization. Emergence offers indeed a bridge between the necessity of complex and adaptive behavior at a macro level (the one of the system) and situationbased ... | [
1628
] | Train |
2,147 | 4 | Reinventing the Familiar: Exploring an Augmented Reality Design Space for Air Traffic Control This paper describes our exploration of a design space for an augmented reality prototype. We began by observing air traffic controllers and their interactions with paper flight strips. We then worked with a multi-disciplinary team of researchers and controllers over a period of a year to brainstorm and prototype ideas for enhancing paper flight strips. We argue that augmented reality is more promising (and simpler to implement) than the current strategies that seek to replace flight strips with keyboard/monitor interfaces. We also argue that an exploration of the design space, with active participation from the controllers, is essential not only for designing particular artifacts, but also for understanding the strengths and limitations of augmented reality in general. Keywords: Augmented Reality, Design Space, Interactive Paper, Participatory Design, Video Prototyping INTRODUCTION Air traffic control is a complex, collaborative activity, with well-established and successful work p... | [
574,
1832,
2301,
2673,
2740,
2756
] | Validation |
2,148 | 1 | Towards Detection of Human Motion Detecting humans in images is a useful application of computer vision. Loose and textured clothing, occlusion and scene clutter make it a difficult problem because bottom-up segmentation and grouping do not always work. We address the problem of detecting humans from their motion pattern in monocular image sequences; extraneous motions and occlusion may be present. We assume that we may not rely on segmentation, nor grouping and that the vision front-end is limited to observing the motion of key points and textured patches in between pairs of frames. We do not assume that we are able to track features for more than two frames. Our method is based on learning an approximate probabilistic model of the joint position and velocity of different body features. Detection is performed by hypothesis testing on the maximum a posteriori estimate of the pose and motion of the body. Our experiments on a dozen of walking sequences indicate that our algorithm is accurate and efficient. | [
1969
] | Test |
2,149 | 3 | Efficient Data and Program Integration Using Binding Patterns : In this work, we investigate data and program integration in a fully distributed peer-to-peer mediation architecture. The challenge in making such a system succeed at a large scale is twofold. First, sharing a resource should be easy; therefore, we need a simple concept for modeling resources. Second, we need an ecient architecture for distributed query execution, capable of handling well costly computations and large data transfers. To model heterogeneous resources, we propose using the unied abstraction of table with binding patterns, simple yet powerful enough to capture data and programs. To exploit a resource with restricted binding patterns, we propose an ecient BindJoin operator, following the classical iterator model, in which we build optimization techniques for minimizing large data transfers and costly computations, and maximizing parallelism. Furthermore, our BindJoin operator can be tuned to deliver most of its output in the early stages of the execution, which is an important asset in a system meant for human interaction. Our preliminary experimental evaluation validates the proposed BindJoin algorithms, and shows they can provide good performance in queries involving distributed data and expensive programs. Key-words: data and program integration, distributed query processing, binding patterns INRIA, Caravel project. Contact: Ioana.Manolescu@inria.fr y PRISM laboratory, University of Versailles and INRIA, Caravel project. Contact: Luc.Bouganim@prism.uvsq.fr z INRIA, Caravel project. Contact: Francoise.Fabret@inria.fr x INRIA, Caravel project. Contact: Eric.Simon@inria.fr Intgration ecace de donnes et de programmes utilisants des patterns d'accs Rsum : Dans ce rapport, nous tudions l'intgration de donnes et de programmes dans une archite... | [
279,
475,
923,
2139
] | Train |
2,150 | 0 | A Framework For Designing, Modeling and Analyzing Agent Based Software Systems The agent paradigm is gaining popularity because it brings intelligence, reasoning and autonomy to software systems. Agents are being used in an increasingly wide variety of applications from simple email filter programs to complex mission control and safety systems. However there appears to be very little work in defining practical software architecture, modeling and analysis tools that can be used by software engineers. This should be contrasted with object-oriented paradigm that is supported by models such as UML and CASE tools that aid during the analysis, design and implementation phases of object-oriented software systems. In our research we are developing a framework and extensions to UML to address this need. Our approach is rooted in the BDI formalism, but stresses the practical software design methods instead of reasoning about agents. In this paper we describe our preliminary ideas Index Terms: Agent-Oriented programming, ObjectOriented programming, BDI, UML 1. | [
580,
1759,
2364
] | Train |
2,151 | 2 | Internet search for Indian languages With the Internet growing at an exponential rate, no single search engine can index all of the web. It is therefore necessary to build specialized search engines that fulfill particular needs of a community of people. An example is Citeseer, which indexes research papers on the web. Also, as the web is increasingly hosting web pages in different languages, it is essential to be able to search for information stored in a specific language. For a search engine aimed at information in a particular language, an easy to use user interface is as essential as good response time and relevance of results. We introduce Shodh, a search engine for an Indian language. A prototype has been developed for selected set of pages and results are satisfactory. The user interface of the search engine includes both querying facilities as well as display of query results in the same language in which the information is stored. Contents 1 | [
488,
2503
] | Validation |
2,152 | 5 | A Visual Interface for Human-Robot Interaction This paper describes the architecture of a human-robot interaction system and recent work on the vision based human-robot interface. The aim of the project is to develop a robotic system that is safely able to work with a human operator. This means in particular that the operator should be allowed into the work space of the robot and even to interact with the robot by direct contact with the manipulator. Safety issues are of major importance here, so open loop force control and backdrivability of the manipulator are mandatory features of the system. The human-robot interface is visionbased to achieve a natural interaction between the operator and the robot. Important aspects of this interaction are the ability of the vision system to find and track the operators face, to recognise facial gestures and to determine the users gaze point. These functions are implemented in a robust way so tracking failures can be compensated for and temporary occlusions of the face are tolerated. 1 Introdu... | [
1428
] | Train |
2,153 | 0 | Active Databases and Agent Systems - A Comparison This paper examines Active Databases and Agent Systems, comparing their purpose, structure, functionality, and implementation. Our presentation is aimed primarily at an audience familiar with active database technology. We show that they draw upon very similar paradigms in their quest to supply reactivity. This presents opportunities for migration of techniques and formalisms between the two fields. 1 fjbailey,kemp,dnk,raog@cs.mu.oz.au 2 georgeff@aaii.oz.au 3 Appears in T.Sellis, editor, Proceedings of the Second International Workshop on Rules in Database Systems, Lecture Notes in Computer Science 985, pages 342-356, Athens, Greece, 1995. 1 Introduction In recent times, two technologies have become prominent in the database and artificial intelligence research communities. An Active Database (ADB) is a system which supplements traditional database functionality by reacting automatically to state changes, both internal and external, without user intervention. An Agent System (A... | [
1113,
1605
] | Train |
2,154 | 2 | Flexible Interoperability in a Federated Digital Library of Theses and Dissertations Federated digital libraries are composed of autonomous, possibly heterogeneous information services distributed across the Internet. Federation provides users with a seamless, integrated view of the collected information. We are creating a federated system for the Networked Digital Library of Theses and Dissertations (NDLTD), an international consortium of universities, libraries, and other supporting institutions focused on electronic theses and dissertations (ETDs). The NDLTD allows its members minimal restrictions and maximal autonomy, so federating requires dealing flexibly with differences among ontologies, data formats, and finding aids involving several thousand ETDs in four formats and two languages. | [
1677
] | Train |
2,155 | 4 | Interacting At a Distance Using Semantic Snarfing . It is difficult to interact with computer displays that are across the room, which can be important in meetings and when controlling computerized devices. A popular approach is to use laser pointers tracked by a camera, but interaction techniques using laser pointers tend to be imprecise, error-prone, and slow. Therefore, we have developed a new interaction style, where the laser pointer (or other pointing technique such pointing with a finger or even eye tracking) indicates the region of interest, and then the item there is copied ("snarfed") to the user's handheld device, such as a Palm or PocketPC handheld. If the content changes on the PC, the handheld's copy will be updated as well. Interactions can be performed on the handheld using familiar direct manipulation techniques, and then the modified version is sent back to the PC. The content often must be reformatted to fit the properties of the handheld to facilitate natural interaction. 1 | [
1780
] | Test |
2,156 | 3 | Querying the Uncertain Position of Moving Objects In this paper we propose a data model for representing moving objects with uncertain positions in database systems. It is called the Moving Objects Spatio-Temporal (MOST) data model. We also propose Future Temporal Logic (FTL) as the query language for the MOST model, and devise an algorithm for processing FTL queries in MOST. 1 Introduction Existing database management systems (DBMS's) are not well equipped to handle continuously changing data, such as the position of moving objects. The reason for this is that in databases, data is assumed to be constant unless it is explicitly modified. For example, if the salary field is 30K, then this salary is assumed to hold (i.e. 30K is returned in response to queries) until explicitly updated. Thus, in order to represent moving objects (e.g. cars) in a database, and answer queries about their position (e.g., How far is the car with license plate RWW860 from the nearest hospital?) the car's position has to be continuously updated. This is unsa... | [
1024,
2476,
2900
] | Train |
2,157 | 1 | Compressive Computation in Analog VLSI Motion Sensors . We introduce several different focal plane analog VLSI motion sensors developed in the past. We show how their pixel-parallel architecture can be used to extract low-dimensional information from a higher dimensional data set. As an example we present an algorithm and corresponding experiments to compute the focus of expansion, focus of contraction and the axis of rotation from natural visual input. A fully integrated system for real-time computation of these quantities is proposed as well. In computer simulations it is shown that the direction of motion vector field is best suited to perform the algorithm even at high noise levels. 1 Analog VLSI Motion Sensors In the past the computer vision communityhas invested much effort into developing motion detection algorithms; for a critical review see [BFB94]. Implementing these algorithms in real-time systems proved challenging for computational reasons. Additionally it has been realized that a motion vector field is useful mainly as star... | [
1889
] | Validation |
2,158 | 4 | Collaborative Representations: Supporting Face to Face and Online Knowledge-building Discourse The present widespread interest in the use of electronic media for presents an unprecedented opportunity for leveraging the computational medium's strengths for learning. However, existing software tools provide only primitive support for online knowledge-building discourse. Further work is needed in supporting coordinated use of disciplinary representations, discourse representations, and knowledge representations. This paper introduces the concept of representational guidance for discourse along with results of an initial study of this phenomenon in face to face situations. The paper then considers the requirements for supporting asynchronous online knowledge-building discourse, finding existing computer mediated communication tools to be particularly deficient in supporting artifact-centered discourse. A solution is proposed that coordinates discourse representations with disciplinary and knowledge representations. 1. Introduction There is a great deal of interest in the use of el... | [
1748
] | Train |
2,159 | 1 | PKDD'98 Tutorial on Scalable, High-Performance Data Mining with Parallel Processing Contents 1 Introduction 2 Overview of 7 different approaches for speeding up data mining in large databases 3 An overview of parallel processing for data mining 4 Parallel rule induction 5 Parallel Instance-Based Learning 6 Parallel Genetic Algorithms 7 Parallel Neural Networks 8 Conclusions Introduction. Problem: How to perform efficient data mining in very large databases. Natural solution: parallelism Performance issues: any sequential data mining algorithm: O(N) parallelism reduces this lower bound to O(N/p) (N = No. of tuples, p = No. of processors) Cost-benefit issues: many data warehouses are already implemented on cost-effective parallel database servers 2 Overview of 7 different approaches for speeding up data mining in large databases. Data-Oriented Approaches: (1) Sampling (reduces number of tuples) (2) Attribute selection (reduces number of attributes) (3) Discretization (reduces number of values of attributes, which in | [
1169
] | Validation |
2,160 | 0 | HEMASL: A Flexible Language to Specify Heterogeneous Agents In the realization of agent-based applications the developer generally needs to use heterogeneous agent architectures, so that each application component can optimally perform its task. Languages that easily model the heterogeneity of agents' architectures are very useful in the early stages of the application development. This paper presents HEMASL, a simple meta-language used to specify heterogeneous agent architectures, and sketches how HEMASL should be implemented in an object-oriented commercial programming language as Java. Moreover, the paper briefly discusses the benefits of adding HEMASL to CaseLP, a LP-based specification and prototyping environment for multi-agent systems, in order to enhance its flexibility and usability. 1. Introduction Intelligent agents and multi-agent systems (MASs) are more and more recognized as the "new" modeling techniques to be used to engineer complex and distributed software applications [12]. Agent-based software engineering is concerned with ... | [
239,
1157,
2364,
3066
] | Train |
2,161 | 2 | Democratic Data Fusion for Information Retrieval Mediators Our research presented in this paper concerns the problem of fusing the results returned by the underlying systems to a mediating retrieval system, also called meta-retrieval system, meta-search engine, or mediator. We propose a fusion technique which is based solely on the actual results returned by each system for each query. The final (fused) ordering of documents is derived by aggregating the orderings of each system in a democratic manner. In addition, the fused ordering is accompanied by a level of democracy (alternatively construed as the level of confidence). | [
1503,
2188
] | Train |
2,162 | 1 | Developing Haptic and Visual Perceptual Categories for Reaching and Grasping with a Humanoid Robot Properties of the human embodiment -- sensorimotor apparatus and neurological structure -- participate directly in the growth and development of cognitive processes against enormous worst case complexity. It is our position that relationships between morphology and perception over time lead to increasingly comprehensive models that describe the agent's relationship to the world. We are applying insight derived from neuroscience, neurology, and developmental psychology to the design of advanced robot architectures. To investigate developmental processes, we have begun to approximate the human sensorimotor configuration and to engage sensory and motor subsystems in developmental sequences. Many such sequences have been documented in studies of infant development, so we intend to bootstrap cognitive structures in robots by emulating some of these growth processes that bear an essential resemblance to the human morphology. In this paper, we will show two related examples in which a humanoid robot determines the models and representations that govern its behavior. The first is a model that captures the dynamics of a haptic exploration of an object with a dextrous robot hand that supports skillful grasping. The second example constructs constellations of visual features to predict relative hand/object postures that lead reliably to haptic utility. The result is a rst step in a trajectory toward associative visual-haptic categories that bounds the incremental complexity of each stage of development. | [
1675,
2812
] | Validation |
2,163 | 2 | Assessment Methods for Information Quality Criteria Information quality (IQ) is one of the most important aspects of information integration on the Internet. Many projects realize and address this fact by gathering and classifying IQ criteria. Hardly ever do the projects address the immense difficulty of assessing scores for the criteria. This task must precede any usage of criteria for qualifying and integrating information. After reviewing previous attempts to classify IQ criteria, in this paper we also classify criteria, but in a new, assessment-oriented way. We identify three sources for IQ scores and thus, three IQ criterion classes, each with different general assessment possibilities. Additionally, for each criterion we give detailed assessment methods. Finally, we consider confidence measures for these methods. Confidence expresses the accuracy, lastingness, and credibility of the individual assessment methods. 1 Introduction Low information quality is one of the most pressing problems for consume rs of information that is di... | [
1252,
2381
] | Train |
2,164 | 0 | Prometheus: A Methodology for Developing Intelligent Agents Abstract. As agents gain acceptance as a technology there is a growing need for practical methods for developing agent applications. This paper presents the Prometheus methodology, which has been developed over several years in collaboration with Agent Oriented Software. The methodology has been taught at industry workshops and university courses. It has proven effective in assisting developers to design, document, and build agent systems. Prometheus differs from existing methodologies in that it is a detailed and complete (start to end) methodology for developing intelligent agents which has evolved out of industrial and pedagogical experience. This paper describes the process and the products of the methodology illustrated by a running example. 1 | [
1049,
1297,
2343
] | Train |
2,165 | 3 | Extending Classical Logic with Inductive Definitions The goal of this paper is to extend classical logic with a generalized notion of inductive definition supporting positive and negative induction, to investigate the properties of this logic, its relationships to other logics in the area of non-monotonic reasoning, logic programming and deductive databases, and to show its application for knowledge representation by giving a typology of definitional knowledge. | [
89
] | Train |
2,166 | 4 | Situated Documentaries: Embedding Multimedia Presentations in the Real World We describe an experimental wearable augmented reality system that enables users to experience hypermedia presentations that are integrated with the actual outdoor locations to which they are are relevant. Our mobile prototype uses a tracked see-through head-worn display to overlay 3D graphics, imagery, and sound on top of the real world, and presents additional, coordinated material on a hand-held pen computer. We have used these facilities to create several situated documentaries that tell the stories of events that took place on our campus. We describe the software and hardware that underly our prototype system and explain the user interface that we have developed for it. | [
172,
468,
1514,
2086,
2549,
2570
] | Train |
2,167 | 1 | An Open Platform for Reconfigurable Control bility. . Openness: Reconfigurability and component interchangeability require software architectures that are flexible and that support tools and algorithms from a variety of sources and domains. This requires a shift away from traditional control system implementation, which tends to be practiced with a particular apJune 2001 IEEE Control Systems Magazine 49 By Linda Wills, Suresh Kannan, Sam Sander, Murat Guler, Bonnie Heck, J.V.R. Prasad, Daniel Schrage, and George Vachtsevanos Wills (linda.wills@ece.gatech.edu), Sander, Guler, Heck, and Vachtsevanos are with the School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0250, U.S.A. Kannan, Prasad, and Schrage are with the School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0250, U.S.A. 1998 Corbis Corp. 0272-1708/01/$10.002001IEEE plication in mind and which makes rigid, | [
973
] | Train |
2,168 | 4 | Workload of a Media-Enhanced Classroom Server We charaterize a workload of media-enhanced classrooms. Such classrooms include equipment for presenting multimedia streams and for capturing streams of information (audio, video and notes) during a lecture. We present detailed quantitative performance measurements of one media-enhanced classroom system, Classroom 2000. We characterize the workload from the point of view of a server that supports multiple classrooms. The workload includes server bandwidth, network bandwidth and server storage requirements. We identify patterns in user behavior, and demonstrate how the number of simultaneous study sessions varies with time of day and with the proximity of a specific date to exams. 1 Workload of a Media-Enhanced Classroom Server 1 Introduction The ways we teach and learn will be dramatically affected by current, unprecedented rates of improvement in computational power and network bandwidth, as well as the development of innovative user interfaces and virtual environments. Alre... | [
2517
] | Train |
2,169 | 4 | Augmented Workspace: The World as Your Desktop . We live in a three dimensional world, and much of what we do and how we interact in the physical world has a strong spatial component. Unfortunately, most of our interaction with the virtual world is two dimensional. We are exploring the extension of the 2D desktop workspace into the 3D physical world, using a stereoscopic see-through head-mounted display. We have built a prototype that enables us to overlay virtual windows on the physical world. This paper describes the Augmented Workspace, which allows a user to pos ition windows in a 3D work area. Keywords. Ubiquitous computing, cooperative buildings, human-computer interaction, physical space, context awareness, visualization. 1. Introduction In our daily lives, much of what we do and how we interact has a strong spatial component. Your calendar is on a wall, or on a certain part of your desk, and sticky notes are placed on walls and whiteboards. Yet, as an increasing portion of our work is done on computers, a large m... | [
571,
762
] | Train |
2,170 | 0 | Exploring auction-based leveled-commitment contracting - Part I: English-type auctioning A key problem addressed in the area of multiagent systems is the automated assignment of multiple tasks to executing agents. The automation of multiagent task assignment requires that the individual agents (i) use a common protocol that prescribes how they have to interact in order to come to an agreement and (ii) fix their final agreement in a contract that specifies the commitments resulting from the assignment on which they agreed. This report describes a novel approach to automated task assignment in multiagent systems that is based on an auction-based protocol and on leveled commitment contracting. This approach is applicable in a broad range of realistic scenarios in which knowledge-intensive negotiation among agents is not feasible and in which future environmental changes may require agents to breach their contracts. 1 Introduction The area of multiagent systems (e.g., [5, 8, 16]), which is concerned with systems composed of technical entities called agents that in... | [
1561,
2568
] | Test |
2,171 | 1 | Theoretical Investigations And Experimental Explorations Of The Necessity Of User Guidance In Case-Based Knowledge Acquisition The intention of the present paper is to justify both theoretically and experimentally that user guidance is inevitable in case-based knowledge acquisition. The methodology of our approach is quite simple: We choose a well-understood area which is tailored to case-based knowledge acquisition. Furthermore, we choose a prototypical case-based learning algorithm which is obviously suitable for the problem domain under consideration. Then, we perform a number of knowledge acquisition experiments. They clearly exhibit essential limitations of knowledge acquisition from randomly chosen cases. As a consequence, we develop scenarios of user guidance. Based on these theoretical concepts, we prove a few theoretical results characterizing the power of our approach. Next, we perform a new series of more constrained results which support our theoretical investigations. The present report aims at presenting a large amount of experimental data exceeding the space available in conference proceedings, ... | [
967,
1844,
2395,
2890,
3160,
3176
] | Validation |
2,172 | 3 | Version Management of XML Documents The problem of ensuring efficient storage and fast retrieval for multi-version structured documents is important because of the recent popularity of XML documents and semistructured information on the web. Traditional document version control systems, e.g. RCS, which model documents as a sequence of lines of text and use the shortest edit script to represent version differences, can be inefficient and they do not preserve the logical structure of the original document. Therefore, we propose a new approach where the structure of the documents is preserved intact, and their sub-objects are timestamped hierarchically for efficient reconstruction of current and past versions. Our technique, called the Usefulness Based Copy Control (UBCC), is geared towards efficient version reconstruction while using small storage overhead. Our analysis and experiments illustrate the effectiveness of the overall approach to version control for structured documents. Moreover UBCC can easily support multiple concurrent versions as well as partial document retrieval. 1 | [
2232,
2524,
2583,
2966
] | Train |
2,173 | 0 | Learning Models of Other Agents Using Influence Diagrams We adopt decision theory as a descriptive paradigm to model rational agents. We use influence diagrams as a modeling representation of agents, which is used to interact with them and to predict their behavior. In this paper, we provide a framework that an agent can use to learn the models of other agents in a multi-agent system (MAS) based on their observed behavior. Since the correct model is usually not known with certainty our agents maintain a number of possible models and assign a probability to each of them being correct. When none of the available models is likely to be correct, we modify one of them to better account for the observed behaviors. The modification refines the parameters of the influence diagram used to model the other agent's capabilities, preferences, or beliefs. The modified model is then allowed to compete with the other models and the probability assigned to it being correct can be arrived at based on how well it predicts the behaviors of the other agent alrea... | [
699,
978
] | Test |
2,174 | 1 | Discriminant Analysis based Feature Extraction We propose a new feature extraction scheme called Discriminant Component Analysis. The new scheme decomposes a signal into orthonormal bases such that for each base there is an eigenvalue representing the discriminatory power of projection in that direction. The bases and eigenvalues are obtained based on certain classification criterion. For simplicity, a criterion used in Fisher's Discriminant Analysis (DA) is chosen and is applied iteratively to implement the scheme. We illustrate the motivation of this new scheme and show how it can be used to construct new distance metrics. We then argue that these new distance metrics are more robust than DA based metrics. Finally, very good classification performance on simulation data and real face images are demonstrated using these new distance metrics. 1 Introduction It is important that for different applications, we use different representations for the same signal [1]. For example, PCA (Principal Component Analysis) or wavelet decompos... | [
2067
] | Train |
2,175 | 1 | Process-Oriented Estimation of Generalization Error Methods to avoid overfitting fall into two broad categories: data-oriented (using separate data for validation) and representation-oriented (penalizing complexity in the model). Both have limitations that are hard to overcome. We argue that fully adequate model evaluation is only possible if the search process by which models are obtained is also taken into account. To this end, we recently proposed a method for process-oriented evaluation (POE), and successfully applied it to rule induction [ Domingos, 1998b ] . However, for the sake of simplicity this treatment made a number of rather artificial assumptions. In this paper the assumptions are removed, and a simple formula for error estimation is obtained. Empirical trials show the new, better-founded form of POE to be as accurate as the previous one, while further reducing theory sizes. 1 Introduction Overfitting avoidance is a central problem in machine learning. If a learner is su#ciently powerful, whatever repre... | [
1979,
3063
] | Train |
2,176 | 2 | Applying Co-Training methods to Statistical Parsing We propose a novel Co-Training method for statistical parsing. The algorithm takes as input a small corpus (9695 sentences) annotated with parse trees, a dictionary of possible lexicalized structures for each word in the training set and a large pool of unlabeled text. The algorithm iteratively labels the entire data set with parse trees. Using empirical results based on parsing the Wall Street Journal corpus we show that training a statistical parser on the combined labeled and unlabeled data strongly outperforms training only on the labeled data. 1 | [
855,
1290,
1366,
1386,
2889,
3119
] | Train |
2,177 | 1 | Advances in Large Margin Classifiers Introduction to Large Margin Classiers Alexander J. Smola GMD FIRST Rudower Chaussee 5 12489 Berlin, Germany smola@rst.gmd.de http://www.rst.gmd.de/smola Peter Bartlett Australian National University, RSISE Canberra, ACT 0200, Australia Peter.Bartlett@anu.edu.au http://keating.anu.edu.au/people/bartlett Bernhard Scholkopf GMD FIRST Rudower Chaussee 5 12489 Berlin, Germany bs@rst.gmd.de http://www.rst.gmd.de/bs Dale Schuurmans Department of Computer Science Unversity of Waterloo Waterloo, Ontario N2L 3G1, Canada dale@cs.uwaterloo.ca http://www.cs.uwaterloo.ca/dale The aim of this chapter is to provide a brief introduction to the basic concepts of large margin classiers for readers unfamiliar with the topic. Moreover it is aimed at establishing a common basis in terms of notation and equations, upon which the subsequent chapters will bu | [
1122,
1400
] | Train |
2,178 | 2 | Using Web Structure for Classifying and Describing Web Pages The structure of the web is increasingly being used to improve organization, search, and analysis of information on the web. For example, Google uses the text in citing documents (documents that link to the target document) for search. We analyze the relative utility of document text, and the text in citing documents near the citation, for classification and description. Results show that the text in citing documents, when available, often has greater discriminative and descriptive power than the text in the target document itself. The combination of evidence from a document and citing documents can improve on either information source alone. Moreover, by ranking words and phrases in the citing documents according to expected entropy loss, we are able to accurately name clusters of web pages, even with very few positive examples. Our results confirm, quantify, and extend previous research using web structure in these areas, introducing new methods for classification and description of pages. | [
159,
301,
980,
1290,
1352,
1591,
2503,
3033,
3077
] | Train |
2,179 | 2 | Co-clustering documents and words using Bipartite Spectral Graph Partitioning Both document clustering and word clustering are important and well-studied problems. By using the vector space model, a document collection may be represented as a word-document matrix. In this paper, we present the novel idea of modeling the document collection as a bipartite graph between documents and words. Using this model, we pose the clustering probliem as a graph partitioning problem and give a new spectral algorithm that simultaneously yields a clustering of documents and words. This co-clustrering algorithm uses the second left and right singular vectors of an appropriately scaled word-document matrix to yield good bipartitionings. In fact, it can be shown that these singular vectors give a real relaxation to the optimal solution of the graph bipartitioning problem. We present several experimental results to verify that the resulting co-clustering algoirhm works well in practice and is robust in the presence of noise. | [
471,
847,
901,
1234,
2324,
2580
] | Validation |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.