MetaData (Ontology) Incremental Building and Refinement Agents

Due to the scattering of information across the Web, information retrieval systems are used to collect information into a common repository to provide quick access to the information and to provide a common format for the data exchange. In addition, these systems can be tailored to support the collection of domain-specific data based on the users\´ interests. Collecting the data to meet these needs based on a specific domain of interests can be challenging. In this paper, we present SBIWS, a semantics based intelligent Web service, to provide information retrieval and integration tailored to the specific interests of the user. The objective of SBIWS research is to create an environment and functionality that provides users with a "What You Get Is What You Need (WYGIWYN)" information service. SBIWS allows users to incrementally build up a domain-specific ontology with attributes that contain semantic metadata which describe the interests, or needs, of the users. This multi-agent system retrieves Web data, as driven by stored metadata, and provides target data for analysis in refining the user\´s domain-specific ontology. This feedback mechanism allows users to gradually build up and to refine the ontology, based on their preferences, through a semi-automatic process