A hybrid Data Mining and Case-Based Reasoning user modelin system architecture

In this paper we present the architecture of a hybrid Data Mining and Case-Based Reasoning system which incorporates a user model to help filter information in order to make it more relevant to the user. The main issue of implementing this hybrid system is a knowledge base which is first derived from a domain information database and where a priori user information is unavailable, the system builds a user preference table by monitoring mouse clicks and updating or refining a rule based user model. The user model and knowledge base are classified using a support vector machine and retrieved from the case based reasoning cycle using a self organizing map. The objective of this project is to combine data mining technology and artificial intelligence pattern classifiers as a means to construct a user oriented Knowledge Base and to link this to the case-based reasoning cycle in order to provide domain specific user relevant information to the user in a timely manner.