Section: Research Program

Knowledge Systems and Semantic Web

Keywords:

knowledge representation, ontology, description logics, classification-based reasoning, case-based reasoning, semantic web, information retrieval

Usually, people try to take advantage of the web by searching for information (navigation, exploration), and by querying documents using search engines (information retrieval). Then people try to analyze the obtained results, a task that may be difficult and tedious. Semantic web is an attempt for guiding search for information with the help of software agents, that are in charge of asking questions, searching for answers, classifying and interpreting the answers. However, a software agent may be able to read, understand, and manipulate information on the web, if and only if the knowledge necessary for achieving those tasks is available, and this is why ontologies are of main importance. Thus, there is a need for knowledge representation languages for annotating documents, describing the content of documents and giving a semantics to this content.

In particular, the knowledge representation language used for designing ontologies is the OWL language, which is based on description logics (DLs [100] ). In OWL, knowledge units are represented within concepts (or classes), with attributes (properties of concepts, or relations, or roles), and individuals. The hierarchical organization of concepts (and relations) relies on a subsumption relation (i.e. a partial ordering).

The inference services are based on subsumption, concept and individual classification, two tasks related to “classification-based reasoning”. Furthermore, classification-based reasoning can be extended into case-based reasoning (CBR), which relies on three main operations: retrieval, adaptation, and memorization. Given a target problem, retrieval consists in searching for a source (memorized) problem similar to the target problem. Then, the solution of the source problem is adapted to fulfill the constraints attached to the target problem, and possibly memorized for further reuse.