A long-lasting research activity aiming at studying methods and techniques for building agents with hybrid case-based architecture specialized in adaptive information filtering and retrieval of information on the Web according to the user needs. Results of this activity cover text categorization, case-based reasoning and focused crawling. Several prototypes have been developed. For example, HOLMES developed in Java with a client/server architecture gives the chance to one or more users to interact from any browser by means of a specific GUI applet. It also includes a focused crawling system able to build local indexes of documents related to a given query/topic.
- CAPUANO N., GAETA M., GASPARETTI F., MICARELLI A.. HOLMES: A Prototype for the Targeted Research of Information about Hi-Tech Companies. In: Abraham et al. (eds.): Computational Intelligence and Applications, pp.233-240, Dynamic Publishers, Atlanta, USA, ISBN: 096403980X, 2002.
- GENTILI G., MARINILLI M., MICARELLI A., SCIARRONE F.. Text Categorization in an Intelligent Agent for Filtering Information on the Web. IJPRAI 15(3): 527-549 (2001)
- MARINILLI M., MICARELLI A., SCIARRONE F.. A Hybrid Case-Based Architecture for Information Filtering on the Web. ICCBR Workshops 1999: 23-32