(for: technical implementers)

Topic Maps
Co-Chairs: Steve Pepper, Founder and Chief Technology Officer, Ontopia AS, Norway; Steven R. Newcomb, Consultant, Coolheads Consulting, USA

Combinatorial Hypermaps vs Topic Maps
Patrice Ossona de Mendez, Researcher, Centre National de la Recherche Scientifique, France
Topic Maps have been introduced to interchangeably represent information about the structure of information resources used to define topics, and the relationships between topics. Graphs and hypergraphs have been introduced in discrete mathematics for a similar reason. Industrial needs related to topic maps not only rely on their representations in a data structure, but also deals with new complex information extraction requests, like Navigation, Structure Analysis, Minimal Cut search, etc. The adapted paradigm seems to be the one of a combinatorial hypermap, which has been introduced some decades ago. This paper presents this combinatorial object, associated data structures and algorithmic considerations about it. Topological graph theory is shown to be the right mathematical and algorithmic tool to handle Topic Maps efficiently.

Graph Clustering in Very Large Topic Maps
Olivier Baudon, Assistant Professor, Labri, UMR 5800, Université Bordeaux1 and Pascal Auillans, Mondeca, France
As topic maps can be modelled by graphs, this presentation will show how graph clustering techniques can be used both to give an overview of large topic maps or in automatic themes generation. Concepts presented in this paper are derived from techniques used in the development of software based on the ISO Topic Map standard.

Data Structures to Support Knowledge Management Systems
Graham D. Moore, Vice President of research and development, empolis GmbH, United Kingdom
In this paper the author questions the suitability of existing data structures and storage mechanisms for providing a useful implementation platform. This paper provides a definitive insight into the properties of knowledge models and how to implement them.

(for: technical implementers)

Topic Maps
Co-Chairs: Steve Pepper, Founder and Chief Technology Officer, Ontopia AS, Norway; Steven R. Newcomb, Consultant, Coolheads Consulting, USA

Harvesting Knowledge from the Organization's Information Assets
Eric D. Freese, Director – Professional Services, ISOGEN International/DataChannel, USA
Most of an organization's corporate knowledge is contained in documents or in the minds of its human resources. To make effective use of this corporate knowledge, organizations must be able to access, harvest, organize and redistribute it. In this session a proof of concept topic map based system with the ability to build and manage topic map documents will be demonstrated. This system has the ability to identify and interpret the information found currently within XML documents, but could be expanded to other document formats. The system can be used to build new topic maps and add into existing ones by reading from any XML document. This is done by simply developing import rules, which interpret the structure of the source document, and building the appropriate topics, associations and occurrences. The information is aggregated to construct and maintain a knowledge base as the document collection grows. The information can also be further enhanced by adding links to other specialized knowledge bases. The system also includes an inferencing engine that allows the user to define rules by which new knowledge can be inferred automatically from the information already known to the system. The rules within the inference engine are themselves topic map structures.

Ferrets and Topic Maps: Knowledge Engineering for an Analytical Engine
James David Mason, Ph.D., Senior Applications Software Engineer, Y-12 National Security Complex, USA
The "Ferret" analytical engine, developed originally by the Y-12 National Security Complex (U.S. Department of Energy) to seek classified data and information associations in documents, combines sophisticated searching with rule-driven analysis and reporting. In its original application, the Ferret engine performs the equivalent of 5000 simultaneous searches and acts on the results, while reading documents at several thousand words per second. The knowledge base that controls the analysis is represented as a series of topic maps. These topic maps encapsulate both a set of hierarchical trees that guide tracing implications of concepts discovered in searching and a set of rules for interpreting implications and initiating actions to be taken when a significant piece of information is found. Because the topic maps can be switched easily, Ferret can be reprogrammed to many tasks, including selection and categorization, scanning of e-mail and newsfeeds, and performing diagnostics and query expansion, in addition to the original classification application.

"tolog" – A Topic Map Query Language
Lars Marius Garshol, Development Manager, Ontopia, Norway
Lars provides a description of a query language for topic maps called tolog, followed by discussions of how to create virtual associations, how to extend it to a full programming language and how to create a natural language processor based on the language.

How Can XML Schemas Enhance Topic Maps?
Martin Bryan, Technical Manager, The Diffuse Project, United Kingdom
XML Schema abstract elements and their associated substitution groups make it possible to tailor XML topic map tags in the same way as you can tailor their SGML counterparts, and allow you to add constraints beyond those that SGML provides.