Discovering the Semantics of Keywords: An Ontology-based Approach∗

In the context of the emerging Semantic Web, a great effort has been done in the construction of ontologies. An increasing number of them is becoming available on the Web, in order to share the knowledge that they represent. In this paper we propose an automatic mechanism that accesses, extracts and semantically merges the knowledge contained in a pool of ontologies available on the Web. In particular, we have focused on the problem of discovering the set of candidate meanings for a given keyword (or keywords). First, the different senses are obtained from different ontology libraries; second, redundant senses are automatically integrated when the system determines that they can be considered synonyms; the previous two steps are repeated until all the ontologies are visited or a specified amount of time is spent. This method proposed to manage keyword senses can be used for different purposes, such as annotation of web pages, ke

Download PDF