Classified advertisement search agent (CASA): a knowledge-based information agent for searching semi-structured text

Information agents are increasingly being used for efficient information retrieval from the Internet. A domain specific agent, Classified Advertisement Search Agent (CASA), has been built for searching real estate advertisements obtained from multiple WWW sites to help users find accommodation. CASA is a knowledge-based system with three main features that distinguish it from other information agents. The first feature is the use of knowledge units (concepts) rather than keywords as the basis for matching. The knowledge unit representation and the text parsing algorithm are particularly useful for information extraction from semi-structured texts such as Web pages generated from online databases. The second feature is the integration of knowledge acquisition with the retrieval process. When CASA executes on the Internet, it can extract suburb names and price statistics to enlarge its knowledge base. The third feature is the information retrieval cycle where feedback from the user and the search result are used to adjust the query before restarting the search.