15th European Conference on Artificial Intelligence
|July 21-26 2002 Lyon France|
We present a new approach to product recommendation that addresses the limitations of the standard case-based reasoning (CBR) approach of retrieving a list of cases that are most similar to a target query. Instead, the target query is used to construct a retrieval network in which cases selected for initial presentation to the user are representative of cases that differ from the target query in similar ways. By following links in the retrieval network, the user can examine alternative solutions with no need to await the retrieval of new cases. Other advantages of the approach include increased diversity among the cases initially presented to the user and the ability to explain why cases are recommended in much the same way as a human salesperson might explain their relevance.
Keywords: Case-Based Reasoning, Information Retrieval
Citation: David McSherry: Recommendation Engineering. In F. van Harmelen (ed.): ECAI2002, Proceedings of the 15th European Conference on Artificial Intelligence, IOS Press, Amsterdam, 2002, pp.86-90.