The opening book is an important component of a chess engine, and thus computer chess programmers have been developing automated methods to improve the quality of their books. For chess, which has a very rich opening theory, large databases of high-quality games can be used as the basis of an opening book, from which statistics relating to move choices from given positions can be collected. In order to find out whether the opening books used by modern chess engines in machine versus machine competitions are ``comparable'' to those used by chess players in human versus human competitions, we carried out analysis on 26 test positions using statistics from two opening books one compiled from humans' games and the other from machines' games. Our analysis using several nonparametric measures, shows that, overall, there is a strong association between humans' and machines' choices of opening moves when using a book to guide their choices.

The k -modes algorithm [1] extends the k -means paradigm to cluster categorical data by using (1) a simple matching dissimilarity measure for categorical objects, (2) modes instead of means for clusters, and (3) a frequency-based method to update modes in the k -means fashion to minimize the cost function of clustering. Because the k -modes algorithm uses the same clustering process as k -means, it preserves the efficiency of the k -means algorithm. Although the k -modes algorithm is very efficient, it suffers the problem that the clustering results are sensitive to the selection of the initial points. Hence, a better initial points selection procedure would improve the reliability and accuracy of clustering results. To that end, an iterative initial-points refinement algorithm for k -modes clustering has been presented in [2]. As shown in [2], the new initialization pr ocedure greatly improves the reliability and accuracy of final clustering results. Despite the su ccess of Ref. [2], the following observations motivate us to further pursue other alternative initialization methods.

The quality of the ranking function is an important factor that determines the quality of the Information Retrieval system. Each document is assigned a score by the ranking function; the score indicates the likelihood of relevance of the document given a query. In the vector space model, the ranking function is defined by a mathematic expression. We propose a fuzzy logic (FL) approach to defining the ranking function. FL provides a convenient way of converting knowledge expressed in a natural language into fuzzy logic rules. The resulting ranking function could be easily viewed, extended, and verified: * if (tf is high) and (idf is high) > (relevance is high); * if (overlap is high) > (relevance is high). By using above FL rules, we are able to achieve performance approximately equal to the state of the art search engine Apache Lucene (deltaP10 +0.92%; deltaMAP -0.1%). The fuzzy logic approach allows combining the logic-based model with the vector model. The resulting model possesses simplicity and formalism of the logic based model, and the flexibility and performance of the vector model.

We are developing a system that aims to perf orm the same inferences as a human reader, on car-crash reports. More precisely, we expect it to determine the causes of the accident as they appear from the text. We describe the genera l semantic framework in which our study takes place, the linguistic and semantic levels of analysis, and the inference rules used by the system.

To understand texts written in natural language (LN), we use our knowledge about the norms of the domain. Norms allow to infer more implicit information from the text. This kind of information can, in general, be defeasible, but it remains useful and acceptable while the text do not contradict it explicitly. In this paper we describe a non-monotonic reasoning system based on the norms of the car crash domain. The system infers the cause of an accident from its textual description. The cause of an accident is seen as the most specific norm which has been violated. The predicates and the rules of the system are stratified: organized on layers in order to obtain an efficient reasoning.

Truth based entailments are not sufficient for a good comprehension of NL. In fact, it can not deduce implicit information necessary to understand a text. On the other hand, norm based entailments are able to reach this goal. This idea was behind the development of Frames (Minsky 75) and Scripts (Schank 77, Schank 79) in the 70's. But these theories are not formalized enough and their adaptation to new situations is far from being obvious. In this paper, we present a reasoning system which uses norms in a causal reasoning process in order to find the cause of an accident from a text describing it.

There has been a lot of recent interest in mining patterns from graphs. Often, the exact structure of the patterns of interest is not known. This happens, for example, when molecular structures are mined to discover fragments useful as features in chemical compound classification task, or when web sites are mined to discover sets of web pages representing logical documents. Such patterns are often generated from a few small subgraphs (cores), according to certain generalization rules (GRs). We call such patterns "generalized patterns"(GPs). While being structurally different, GPs often perform the same function in the network. Previously proposed approaches to mining GPs either assumed that the cores and the GRs are given, or that all interesting GPs are frequent. These are strong assumptions, which often do not hold in practical applications. In this paper, we propose an approach to mining GPs that is free from the above assumptions. Given a small number of GPs selected by the user, our algorithm discovers all GPs similar to the user examples. First, a machine learning-style approach is used to find the cores. Second, generalizations of the cores in the graph are computed to identify GPs. Evaluation on synthetic data, generated using real cores and GRs from biological and web domains, demonstrates effectiveness of our approach.

Event-driven reactive functionalities are an urgent need in nowadays distributed service-oriented applications and (Semantic) Web-based environments. An important problem to be addressed is how to correctly and efficiently capture and process the event-based behavioral, reactive logic represented as ECA rules in combination with other conditional decision logic which is represented as derivation rules. In this paper we elaborate on a homogeneous integration approach which combines derivation rules, reaction rules (ECA rules) and other rule types such as integrity constraint into the general framework of logic programming. The developed ECA-LP language provides expressive features such as ID-based updates with support for external and self-updates of the intensional and extensional knowledge, transactions including integrity testing and an event algebra to define and process complex events and actions based on a novel interval-based Event Calculus variant.

This paper addresses the problem of building a long-living a utonomous agent; by long-living, we mean that this agent has a large number of relatively complex and varying tasks to perform. Biology sugge sts some ideas about the way animals deal with a variety of tasks: brains are made of specialized and complementary areas/modules; skills are spre ad over modules. On the one hand, distributing functions and representation s has immediate advantages: parallel processing implies reaction speed-u p; a relative independence between modules gives more robustness. Both prope rties might clearly increase the agent's efficiency. On the other hand, th e fact of distributing a system raises a fundamental issue: how does the o rganization process of the modules happen during the life-time? 1 There has been much research about the design of modular inte lligent architectures (see for instance [15] [5] [1] [7]). It is neve rtheless very often the (human) designer who decides the way modules are connect ed to each other and how they behave with respect to the others.

I t is stri ctly based on De scriptio n Logic. In addition to c la ss desc ri p-tion of par ame te rs it a lso a llo w s t he modelling of r el at ions be twe en pa rame - t e rs a nd t he prec ise de sc ript i on of the size of da ta to be supplied to a se r vic e. In parti cul a r, i t sol ves two ma jor is sues identi fie d withi n curre nt prop osals f or a Se manti c Web S e rvi ce annotatio n st anda rd. This shall be achi eve d by semantically annot ating web reso ur ces, i.e. el ements of the re sources are as so ci at ed with elements of onto logies. Under the auspices of the W 3C, a couple of language s tandards have bee n devel oped and widely accep te d: buil din g on the Resour ce Description Framework, RDF, [11], and th e RDF Sc hema language, [ 12], concepts and con cep t relations can be def ined in the W eb Ontol ogy Language OW L, [7 ]. However, the web also offers a ccess t o se rvices, i.e., W eb Ser vices.