Toward agency and ontology for web-based information retrieval

This paper presents a system that selects information sources, browses websites, and monitors changes in selected websites. Its distinguishing features are 1) adopting a holistic approach for bolstering web-based information gathering using a society of agents and 2) exploiting features of ontology and lexical chain for processing user queries and filtering relevant information. By engineering web-based information gathering systems along the dimensions of agency and ontology, relations among concepts can be identified and established, and information filtering and monitoring tasks can be automated by information filtering agents (IFAs) and information monitoring agents (IMAs). IFAs and IMAs relieve users of the tasks of browsing and monitoring changes in websites and multiple web sites can be browsed and monitored in parallel by several IFAs and IMAs. By establishing ontological relations among (key)words 1) the query processing agent (QPA) selects appropriate URLs for a query and 2) IFAs search and filter information relevant to the query. Empirical results demonstrated that the QPA can find appropriate number of websites and IFAs are effective in filtering relevant information.