Search

Information retrieval (IR) deals with representation, organization, and storage of information. IR should provide the user with an easy access to the information in which he or she is interested. However, characterization of the user information needs is a complex problem. Usually, a full description of the user information needs cannot be made directly and a set of keywords should rather be used. Thus, given a set of keywords, an IR system retrieves information, which may be relevant to the user. However, traditional IR systems overload recovery information. Personalization systems, in particular, the recommender ones, have been proposed in order to mitigate the impact of an enormous amount of retrieved information on the user. This paper describes the architecture of PREDICA, an experimental platform for combining advanced query and recommender systems, adaptive interface, data warehouse and knowledge discovery on database in a digital library. PREDICA uses a pre-processing strategy for improving the search of information and personalization strategies for documents' recommendation. Among other open source digital library research projects, like D-Space, Eprints, and Greenstone, PREDICA has an integrated recommender system to provide helpful information to the users. [ABSTRACT FROM AUTHOR]