A Model for Information Retrieval Agent System Based on Keywords Distribution

Information retrieval is one of the most important technologies at present. We can always get many information in the Internet or distributed computing systems using various information retrieval models. For searching proper information that we need, it is necessary to construct efficient information retrieval agent systems helping many Web clients' requests. In this paper, we propose a simple new model for information retrieval agents based on many terms or keywords distribution in a document or distributed database. For the key paragraph extraction we use meaningful term's frequency and the key word distribution characteristics in a document, and those terms are selected by using stemming, filtering stop-lists, synonym for search meaningful terms in a document. The agent receives a Web client's information retrieval request and extracts key paragraph with frequency and distribution using the keywords of the client, and then the agent constructs profile of the documents with the keywords, key paragraph, address of the document browsing. And then we can search many documents or knowledge easily using the profile for information retrieval and browse the document.