The Wifs system is capable of filtering HTML or text documents retrieved by the search engine AltaVista in response to a query input by the user. This system evaluates and reorders page links returned by the search engine, taking into account the user model of the user who typed in the query. The user can provide feedback on the viewed documents, and the system uses that feedback to update the user model accordingly. The filtering system is based on a content-based approach, where the documents retrieved by the search engine are assessed solely according to their contents. User models consist of a frame whose slots contain terms (topics), each one associated with other terms (co-keywords) which form a simple semantic network. Slot terms, that is, the topics, must be selected from those contained in a Terms Data Base, created a priori by experts who select the terms deemed most relevant for the pertinent domain.
- MICARELLI A., SCIARRONE F.. Anatomy and Empirical Evaluation of an Adaptive Web-Based Information Filtering System. User Model. User-Adapt. Interact. 14(2-3): 159-200 (2004).