Cluster-based adaptive information retrieval

The paper discusses the issues involved in the design of a complete information retrieval system based on user-oriented clustering schemes. Clusters are constructed taking into account the users´ perception of similarity between documents. The system accumulates feed-back from the users and employs it to construct user-oriented clusters. An optimization function to improve the effectiveness of the clustering process is developed. A retrieval process based on the clustering scheme is described. The system developed is experimentally validated and compared with existing systems