Fuzzy Bayesian Classifier: a Multi-Agent System for Information Retrieval in the Web

A fuzzy Bayesian approach helping the Internet user to filter Web pages is discussed. In the proposed approach, one page can be classified as having the continuous quality of being interesting, this means that a certain grade of membership can be associated with each page relative to a category of selection. Filtering is based on the evidences of the content of the page title, abstract or complete document. An example comparing crisp and fuzzy classifiers implemented as a part of multi-agent system to support information filtering and retrieval in the Web is discussed illustrating the proposed approach.