Tutorials |
Link Mining: Current State of the Art |
Ronen Feldman |
Bar-Ilan University, Israel |
Email: |
Building an intelligent Web: theory and practice |
Pawan Lingras and Rajendra Akerkar |
Saint Mary's University, Halifax, Canada |
Email: Pawan Lingras () |
Towards Semantic Service-Oriented Systems on the Web |
Sung-Kook Han 1 and Dumitru Roman 2 |
1 Won Kwang University, Korea and 2 DERI Innsbruck, Austria |
Email: Dumitru Roman () |
Knowledge extraction for improving agent efficiency |
Andreas L. Symeonidis and Pericles A. Mitkas |
Aristotle University of Thessaloniki, Greece |
Email: Andreas Symeonidis () |
Abstracts |
Link Mining: Current State of the Art
Ronen Feldman
Bar-Ilan University, Israel
Abstract
The information age has made it easy to store large amounts of data. The proliferation of documents available on the Web, on corporate intranets, on news wires, and elsewhere is overwhelming. However, while the amount of data available to us is constantly increasing, our ability to absorb and process this information remains constant. Search engines only exacerbate the problem by making more and more documents available in a matter of a few key strokes. Link Mining is a new and exciting research area that tries to solve the information overload problem by using techniques from data mining, machine learning, Information Extraction, Text Categorization, Visualization and Knowledge Management. Link Mining is the process of building up networks of interconnected objects through various relationships in order to discover patterns and trends. The main tasks of Link Mining are to extract, discover, and link together sparse evidence from vast amounts of data sources, to represent and evaluate the significance of the related evidence, and to learn patterns to guide the extraction, discovery, and linkage of entities. The relationships could be transactional, geographical, social, or temporal. Link Mining involves the preprocessing of document collections (text categorization, term extraction, and information extraction), integration with structured information sources, the storage of the intermediate representations, the techniques to analyze these intermediate representations (distribution analysis, clustering, trend analysis, association rules, etc.) and visualization of the results. In this tutorial we will present the general theory of Link Mining and will demonstrate several systems that use these principles to enable interactive exploration of a combination of structured and unstructured collections. We will present a general architecture of Link Mining systems that operate on the web and will outline the algorithms and data structures behind the systems. The Tutorial will cover the state of the art in this rapidly growing area of research. Several real world applications of Link Mining will be presented.
Building an intelligent Web: theory and practice
Pawan Lingras and Rajendra Akerkar
Saint Mary's University, Halifax, Canada.
Abstract
Web mining draws from a wide range of techniques derived from Artificial Intelligence (AI) and Information Retrieval (IR). The application of these AI and IR techniques can be very useful in development of the next generation of intelligent Web sites - the web sites that adapt to users' information needs. This is one of the most active and exciting research areas. Many researchers in areas such as artificial intelligence, data visualization, statistics, and machine learning are contributing to this field. The breadth of the field makes it difficult to grasp its development. The objective of this tutorial is to present Web intelligence techniques in an organized, comprehensive, in-depth, yet very lucid manner that is accessible to students, researchers and a wide range of Web Technology developers. Tutorial will describe how theoretical bases, can be illustrated with the help of simple numeric examples, followed by practical implementations. The topics that will be covered include:
- Information retrieval
- Semantic Web
- Classification and association
- Clustering
- Web usage mining
- Web content mining
- Web structure mining
Teachers will find it useful for teaching a course on Web Mining, Web Intelligence or Web Information Retrieval at an advanced undergraduate level or first year graduate level. Researchers will be able to use the overview to explore new research projects.
Towards Semantic Service-Oriented Systems on the Web
Sung-Kook Han 1 and Dumitru Roman 2
1 Won Kwang University, Korea and 2 DERI Innsbruck, Austria
Abstract
A new paradigm - service-orientation - is emerging nowadays for distributed computing and e-business processing; this new paradigm utilizes "services" (autonomous platform-independent computational elements that can be described, published, discovered and accessed over the Internet using standard protocols) as fundamental elements for developing applications/solutions. In recent years, various forms of service-oriented metaphors have appeared; amongst them, Web services, Grid services, Semantic Web Services, and e-Services are the most important. Recent research (to which we generally refer to as Semantic Service-Oriented Systems), which draws on a variety of fields such as Semantic Web, Web Services, knowledge representation, formal methods, software engineering, process modelling, and software agents, is gaining momentum. Research in the mentioned fields can be exploited to automate Web services-related tasks, like discovery, selection, composition, mediation, monitoring, and invocation, thus enabling seamless interoperation between them while keeping human intervention to a minimum.
In this context, the objectives of this tutorial are to give an overview of the aims and challenges of Semantic Service-Oriented Systems on the Web, to explain concepts, models, languages and technologies that enable semantics in the context of Service-Oriented Systems on the Web, as well as discussing recent advances in semantics for services on the Web. The tutorial is designed to provide attendees with the required information and knowledge, enabling them to use Semantic Service-Oriented on the Web and to evaluate upcoming systems and technologies in a better way.
The tutorial targets academics, industrial researchers, and developers interested in developing the next generation of services on the Web. Although no specific knowledge is demanded as a pre-requisite for attending the tutorial, basic knowledge about the Web, Semantic Web, ontologies, and Service-oriented Architectures will allow attendees to better understand and follow the tutorial.
The tutorial will be organized in two main parts: in the first part the attendees will gain a detailed understanding of the aims and challenges of Semantically Service-oriented Systems on the Web, the main concepts they define, the languages and infrastructures used, as well as the most recent approaches in the area; the second part of the tutorial will focus on a specific use case and will show how a specific technology (WSMO/WSML/WSMX) tackles the challenges.
Knowledge extraction for improving agent efficiency
Andreas L. Symeonidis and Pericles A. Mitkas
Aristotle University of Thessaloniki, Greece
Abstract
The tutorial will offer a self-contained overview of a relatively young but important area of research: the intersection of agent intelligence and data mining. This intersection leads to considerable advancements in the area of information technologies, drawing the increasing attention of both research and industrial communities. It can take two forms: a) the more mundane use of intelligent agents for improved data mining and, b) the use of knowledge extraction techniques for smarter, more efficient agents. The second approach is the main focus of the tutorial.
The tutorial will provide a review of data mining and agent technology fields, and provide a number of approaches to the problem of infusing knowledge extracted by the use of data mining techniques, into agents. It shall elaborate on a methodology for developing multi-agent systems, describe available open-source tools to support this process, and demonstrate the application of the methodology on different types of MAS.
Specific objectives of the tutorial include:
- The familiarization of researchers and developers with the idea of infusing inductive logic into deductive systems
- The presentation of all issues concerning the coupling of data mining and intelligent agent technologies
- The presentation of all state-of-the-art tools and methodologies for embedding knowledge into agents
- The presentation of novel data mining techniques for knowledge extraction on intelligent agents’ behavioral characteristics
- The presentation of a number of different test case scenarios that abide by (and take advantage of) the proposed coupling.