Bücher der Reihe Lecture Notes in Artificial Intelligence

Peer-to-peer (P2P) computing is currently attracting enormous public attention, spurred by the popularity of file-sharing systems such as Napster, Gnutella, Morpheus, Kaza, and several others. In P2P systems, a very large number of autonomous computing nodes, the peers, rely on each other for services. P2P networks are emerging as a new distributed computing paradigm because of their potential to harness the computing power and the storage capacity of the hosts composing the network, and because they realize a completely open decentralized environment where everybody can join in autonomously. Although researchers working on distributed computing, multiagent systems, databases, and networks have been using similar concepts for a long time, it is only recently that papers motivated by the current P2P paradigm have started appearing in high quality conferences and workshops. In particular, research on agent systems appears to be most relevant because multiagent systems have always been thought of as networks of autonomous peers since their inception. Agents, which can be superimposed on the P2P architecture, embody the description of task environments, decision-support capabilities, social behaviors, trust and reputation, and interaction protocols among peers. The emphasis on decentralization, autonomy, ease, and speed of growth that gives P2P its advantages also leads to significant potential problems. Most prominent among these are coordination - the ability of an agent to make decisions on its own actions in the context of activities of other agents, and scalability - the value of the P2P systems in how well they self-organize so as to scale along several dimensions, including complexity, heterogeneity of peers, robustness, traffic redistribution, etc.This book brings together an introduction, three invited articles, and revised versions of the papers presented at the Second International Workshop on Agents and Peer-to-Peer Computing, AP2PC 2003, held in Melbourne, Australia, July 2003.

Since 1993, PROPOR Workshops have become an important forum for - searchers involved in the Computational Processing of Portuguese,both written and spoken. This PROPOR Workshop follows previous workshops held in 1993 (Lisbon, Portugal), 1996 (Curitiba, Brazil), 1998 (Porto Alegre, Brazil), 1999 ' (Evora, Portugal), 2000 (Atibaia, Brazil) and 2003 (Faro, Portugal). The wo- shop has increasingly contributed to bringing together researchers and industry partners from both sides of the Atlantic. The constitution of an international Program Committee and the adoption of high-standard referee procedures demonstrate the steady development of the ?eld and of its scienti?c community. In 2006 PROPOR received 56 paper submissions from 11 di?erent countries: Brazil, Portugal, Spain, Norway, USA, Italy, Japan, France, Canada, Denmark and the UK, from which 9 are represented in the accepted papers. Each submitted paper underwent a careful, triple-blind review by the P- gram Committee. All those who contributed are mentioned in the following pages. The reviewing process led to the selection of 20 regular papers for oral presentation and 17 short papers for poster sections, which are published in this volume. The workshop and this book were structured around the following main t- ics, seven for full papers: (i) automatic summarization; (ii) resources; (iii) au- matic translation; (iv) named entity recognition; (v) tools and frameworks; (vi) systems and models; and another ?ve topics for short papers; (vii) information extraction; (viii) speech processing; (ix) lexicon; (x) morpho-syntactic studies; (xi) web, corpus and evaluation.

ESAW 2005 took place at the Pine Bay Hotel in Kusadasi, Turkey at the end of the October 2005. It was organized as a stand-alone event as were ESAW 2004and ESAW 2003.Following the initial ESAW vision,which wasset in 1999, by the members of the working group on "e;Communication, Coordination and Collaboration"e; of Agentlink, ESAW 2005 continued to focus on the engineering of complex software systems in terms of multi-agent societies, especially the social and environmental aspects of such societies. The number of participants (40 researchers from 10 countries) and the highly interactive discussions held during the workshop showed the augmented importance of the initial vision as well as the e?ectiveness of ESAW as a well-established research forum. It is obvious that today's interconnected world increases the importance of approaches concerning the engineering of complex and distributed software systems. These kinds of large scale systems, made up of massive numbers of autonomous components, force us to discover new and novel approaches to model and engineer such systems as agent societies. It is very likely that such innovations will exploit lessons from a variety of di?erent scienti?c disciplines, such as sociology, economics, organizational science and biology: ESAW 2005 included presentations from these domains in addition to its traditional - search topics.

This book constitutes the refereed proceedings of the 19th Annual Conference on Learning Theory, COLT 2006, held in Pittsburgh, Pennsylvania, USA, June 2006. The book presents 43 revised full papers together with 2 articles on open problems and 3 invited lectures. The papers cover a wide range of topics including clustering, un- and semi-supervised learning, statistical learning theory, regularized learning and kernel methods, query learning and teaching, inductive inference, and more.

The area of smart homes is fast developing as an emergent area which attracts the synergy of several areas of science. This volume offers a collection of contributions addressing how artificial intelligence (AI), one of the core areas of computer science, can bring the growing area of smart homes to a higher level of functionality where homes can truly realize the long standing dream of proactively helping their inhabitants in an intelligent way.After an introductory section to describe a smart home scenario and to provide some basic terminology, the following 9 sections turn special attention to a particular exemplar application scenario (provision of healthcare and safety related services to increase the quality of life) exploring the application of specific areas of AI to this scenario.

th The 14 International Conference on Conceptual Structures (ICCS 2006) was held in Aalborg, Denmark during July 16 - 21, 2006. Responding to the Call for Papers, we received 62 papers from 20 different countries, representing six different continents. This clearly indicates the international nature of the ICCS community as well as the widespread interest which was spawned by the previous conferences. By a thorough review process, 24 papers were selected to be included in this volume. In addition, six invited speakers made contributions which can be found in the first section of this volume. The theme of ICCS 2006--Conceptual Structures: Inspiration and Application-- points to a dual focus of interest that is also reflected in the constellation of papers. From the beginning of the planning of this conference, we focused on inspirational sources that have led to the current state of research in our community, by tracing important historical influences which daily effect work in representing knowledge and in handling representations of conceptual structures. At the same time, we also focused on ways in which these legacies are employed to further advance theory and practice in the field of knowledge representation and processing. With this volume, we believe that a valuable contribution to both aspects of this field is being made. We wish to express our appreciation to all the authors of submitted papers, to the members of the Editorial Board and the Program Committee for all their work and valuable comments.

The Turkish Artificial Intelligence and Neural Network Symposium (TAINN) is an annual meeting where scientists present their new ideas and algorithms on artificial intelligence and neural networks with either oral or poster presentation. The TAINN- Turkish Conference on AI and NN Series started in 1992 at Bilkent University in Ankara, envisioned by various researchers in AI and NN then at the Bilkent, Middle East Technical, Bo azici and Ege universities as a forum for local researchers to get together and communicate. Since then, TAINN has been held annually around early summer. This year the 14th TAINN conference was organized by the EE and CE departments of the zmir Institute of Technology with an emphasis on international contributions. Among the 75 papers, 41 were accepted for oral presentation and 12 for poster presen-tation. In addition to the presentations, invited lectures were given by Burhan Turk en (Knowledge/Intelligence Systems Laboratory, University of Toronto) and Joerg Siekmann (German Research Center for Artificial Intelligence DFKI, Saarland University), providing us with new developments in the field. We are very grateful to the contributions of such pioneer scientists.

Here are the refereed proceedings of the First International Conference on Knowledge Science, Engineering and Management, KSEM 2006, held in Guilin, China in August 2006 in conjunction with PRICAI 2006. The book presents 51 revised full papers and 57 revised short papers together with 4 invited talks, reporting a wealth of new ideas and current research results in the broad areas of knowledge science, knowledge engineering, and knowledge management.

This book constitutes the refereed proceedings of the 8th European Conference on Case-Based Reasoning, ECCBR 2004, held in Fethiye, Turkey in September 2006. The book presents 31 revised full papers and 5 revised application papers together with 2 invited papers and 2 abstracts of invited talks. The coverage represents snapshot of current current issues in case-based reasoning, ranging from theoretical and methodological issues to advanced applications in various fields.

This book constitutes the refereed proceedings of the 10th European Conference on Logics in Artificial Intelligence, JELIA 2006. The 34 revised full papers and 12 revised tool description papers presented together with 3 invited talks were carefully reviewed and selected from 96 submissions. The papers cover a range of topics within the remit of the Conference, such as logic programming, description logics, non-monotonic reasoning, agent theories, automated reasoning, and machine learning.

Here are the refereed proceedings of the 8th International Conference on Artificial Intelligence and Symbolic Computation, AISC 2006.The book presents 18 revised full papers together with 4 invited papers, organized in topical sections on artificial intelligence and theorem proving, symbolic computation, constraint satisfaction/solving, and mathematical knowledge management. Coverage includes mathematical foundations, implementations, and applications in industry and academia.

This book constitutes the refereed proceedings of the Third International Workshop on the Emergence and Evolution of Linguistic Communication, EELC 2006. The book presents 12 revised full papers together with 5 invited papers. These focus on the evolution and emergence of language - a fast growing interdisciplinary research area touching such different disciplines as anthropology, linguistics, psychology, primatology, neuroscience, cognitive science and computer science.

Thisbookcontainsthepostworkshopproceedingsofthe7thInternationalWo- shop on Knowledge Discovery from the Web, WEBKDD 2005. The WEBKDD workshop series takes place as part of the ACM SIGKDD International Conf- ence on Knowledge Discovery and Data Mining (KDD) since 1999. The discipline of data mining delivers methodologies and tools for the an- ysis of large data volumes and the extraction of comprehensible and non-trivial insights from them. Web mining, a much younger discipline, concentrates on the analysisofdata pertinentto theWeb.Web mining methods areappliedonusage data and Web site content; they strive to improve our understanding of how the Web is used, to enhance usability and to promote mutual satisfaction between e-business venues and their potential customers. In the last years, the interest for the Web as medium for communication, interaction and business has led to new challenges and to intensive, dedicated research. Many of the infancy problems in Web mining have now been solved but the tremendous potential for new and improved uses, as well as misuses, of the Web are leading to new challenges.

th The 15 International Conference on Knowledge Engineering and Knowledge Management (2006), held on October 6-10, 2006 in Pod? ebrady,Czech Republic, followed a long tradition of European Knowledge Acquisition Workshops (from 1987),whicheventuallyacquiredtheformatofconference(in2000)whilekeeping their open-minded and interactive spirit. During the nearly 20 year lifespan of the series, the discipline of knowledge engineering (KE) evolved greatly. While knowledge acquisition (KA) techniques dominated in the very ?rst years, formal approaches to knowledge-based inf- ence and variousnew streamssuch as knowledgediscoveryfrom data/textslater cameinto play. During thelate1990sandafterwards,EKAWbecamea founding community for ontology and Semantic Web Research, which was also re?ected in the sub-titles of the 2002 and 2004 editions: "e;Ontologies and the Semantic Web"e; and "e;Engineering Knowledge in the Age of the Semantic Web,"e; resp- tively. The 2006 edition, in turn, only slightly refocussed this trend. Its sub-title is "e;Managing Knowledge in a World of Networks,"e; which re?ects the fact that semantics typically arises not only as a result of explicit engineering activities (as in Semantic Web) but also emerges from interaction of a high number of interconnected documents, ontological concepts, software applications and - especially - human users. The importance given to the interconnection of - man users in a sense loops back to the knowledge acquisition roots of EKAW and its 'holistic' view of knowledge engineering.

Finding knowledge - or meaning - in data is the goal of every knowledge d- covery e?ort. Subsequent goals and questions regarding this knowledge di?er amongknowledgediscovery(KD) projectsandapproaches. Onecentralquestion is whether and to what extent the meaning extracted from the data is expressed in a formal way that allows not only humans but also machines to understand and re-use it, i. e. , whether the semantics are formal semantics. Conversely, the input to KD processes di?ers between KD projects and approaches. One central questioniswhetherthebackgroundknowledge,businessunderstanding,etc. that the analyst employs to improve the results of KD is a set of natural-language statements, a theory in a formal language, or somewhere in between. Also, the data that are being mined can be more or less structured and/or accompanied by formal semantics. These questions must be asked in every KD e?ort. Nowhere may they be more pertinent, however, than in KD from Web data ("e;Web mining"e;). This is due especially to the vast amounts and heterogeneity of data and ba- ground knowledge available for Web mining (content, link structure, and - age), and to the re-use of background knowledge and KD results over the Web as a global knowledge repository and activity space. In addition, the (Sem- tic) Web can serve as a publishing space for the results of knowledge discovery from other resources, especially if the whole process is underpinned by common ontologies.

Since knowledge was recognized as a crucial part of intelligent systems in the 1970s and early 1980s, the problem of the systematic and efficient acquisition of knowledge was an important research problem. In the early days of expert systems, the focus of knowledge acquisition was to design a suitable knowledge base for the problem - main by eliciting the knowledge from available experts before the system was c- pleted and deployed. Over the years, alternative approaches were developed, such as incremental approaches which would build a provisional knowledge base initially and would improve the knowledge base while the system was used in practice. Other approaches sought to build knowledge bases fully automatically by employing machine-learning methods. In recent years, a significant interest developed regarding the problem of constructing ontologies. Of particular interest have been ontologies that could be re-used in a number of ways and could possibly be shared across diff- ent users as well as domains. The Pacific Knowledge Acquisition Workshops (PKAW) have a long tradition in providing a forum for researchers to exchange the latest ideas on the topic. Parti- pants come from all over the world but with a focus on the Pacific Rim region. PKAW is one of three international knowledge acquisition workshop series held in the Pacific-Rim, Canada and Europe over the last two decades. The previous Pacific Knowledge Acquisition Workshop, PKAW 2004, had a strong emphasis on inc- mental knowledge acquisition, machine learning, neural networks and data mining.

Knowledge means power - but only if it is available at the right time, the right place, and in the hands of the right people. Structured, engineered, repeatable methodsto gather,transport,andapplyknowledgearecollectivelycalledkno- edge management. Declarative programming strives for the ideal of programming by wish: the user states what he or she wants, and the computer ?gures out how to achieve it. Thus, declarative programming splits into two separate parts: methods for humans on how to write wishes, and algorithms for computers that ful?l these wishes. Knowledgemanagementisnowrecognizedasaneconomickeyfactor.Decl- ative programming has matured far beyond the research stage of a merely - teresting formal logic model to one of the powerful tools in computer science. Nowadays,no professionalactivity isthinkable without knowledgemanagement, and companies increasingly need to document their business processes. Here, declarative programming carries the promise to be a shortcut to not only do- menting but also implementing knowledge-based enterprises. This volume presents a selection of papers presented at the 16th Inter- tional Conference on Applications of Declarative Programming and Knowledge Management, INAP 2005,held in October 2005 at Waseda University, Fukuoka, Japan. These papers re?ect a snapshot of ongoing research and current app- cations in knowledge management and declarative programming. Further, they provide reality checks and many pointers for readers who consider introducing related technologies into their products or working environments. Skimming through the table of contents, technology managers as well as - plementorswillbesurprisedonthewidescopecoveredbythisselectionofpapers. If you think of knowledge streams as supply, manufacturing, or distribution chains, you will see that it all ?ts together.

This book is a follow-up of LNCS volume 2929 with the same title, and presents the major results of COST action 274 (2002-2005), TARSKI: Theory and - plications of Relational Structures as Knowledge Instruments. Relational structures abound in the daily environment: relational databases, data-mining, scaling procedures, preference relations, etc. Reasoning about, and with, relations has a long-standing European tradition, which may be divided into three broad areas: 1. Algebraic Logic: algebras of relations, relational semantics, and algebras and logics derived from information systems. 2. Computational Aspects of Automated Relational Reasoning: decidability and complexity of algorithms, network satisfaction. 3. Applications: social choice, AI, linguistics, psychology, economics, etc. The main objective of the ?rst TARSKI book (LNCS 2929) was to advance the understanding of relational structures and the use of relational methods in applicable object domains. There were the following sub-objectives: 1. Tostudythesemanticalandsyntacticalaspectsofrelationalstructuresarising from 'real world' situations 2. To investigateautomatedinference for relationalsystems, and, wherepossible or feasible, develop deductive systems which can be implemented into industrial applications, such as diagnostic systems 3. To develop non-invasive scaling methods for predicting relational data 4. To make software for dealing with relational systems commonly available We are con?dent that the present book will further the understanding of int- disciplinary issues involving relational reasoning. This book consists of papers which give a clear and self-contained overview of the results obtained by the TARSKI action, typically obtained by di?erent persons from di?erent work - eas.

The papers in this volume were presented at TMRA 2006, the International Conference on Topic Maps Research and Applications, held October 11-12, 2006, in Leipzig, Germany. TMRA 2006 was the second conference of an annual series of international conferences dedicated to Topic Maps in research and industry. Topic maps are continuously gaining more and more attention in science and ind- try; they are "e;crossing the chasm."e; The TMRA series provides a platform for resear- ers in the topic maps community to meet for exciting exchanges in a stimulating setting. The uniqueness of TMRA is its focus on both sides of the same coin: scientific research on topic maps and upcoming applications in industry. In the autumn of 2005 the first TMRA conference took place in Leipzig. The p- ceedings have been published in this LNAI series as volume 3873. It was amazing to see how ideas and solutions from TMRA 2005 matured within the last year to full products or projects. The overall success of TMRA 2005 encouraged us to improve this conference series for the topic maps community: More people were attracted, and the scientific quality was enhanced. The TMRA 2006 program attracted a very international crowd of about 80 att- dees, hosted in the completely new media campus of the Leipzig Media Foundation.

This book constitutes the refereed proceedings of the 20th Annual Conference on Learning Theory, COLT 2007, held in San Diego, CA, USA in June 2007. It covers unsupervised, semisupervised and active learning, statistical learning theory, inductive inference, regularized learning, kernel methods, SVM, online and reinforcement learning, learning algorithms and limitations on learning, dimensionality reduction, as well as open problems.