Modeling, Extracting and Visualizing an Organization's Knowledge with Topic Maps

This paper presents the results of our organization (university) modeling and knowledge extraction from the university homcpages and a pamphlet. The ISO/1EC 13250 standard topic maps are used during the modeling. Tasks concerning topics, topic types, associations, roles and occurrences, definitions and extraction from the existing documents (in HTML or PDF file) were the primary focus. Although ontology and database schema exist in the university domain, the extraction of topics within documents is very difficult when applying those metadata to our specific university. We model the university from our understanding and methodology. The result is a metamodel, which is visualized with the tool. It can be concluded from this experiment that the student easily learns topic maps rather than other metadata modeling such as RDF or OWL. Constrained metamodel languages are easily manipulated rather than using unlimited online ontology.