Information Sources Selection Methodology for Recommender Systems Based on Intrinsic Characteristics and Trust Measure

SILVANA ACIAR; JOSEP LLUIS DE LA ROSA I ESTEVA; JOSEFINA LÓPEZ HERRERA

Vancouver, British Columbia, Canada

Workshop; Intelligent Techniques for Web Personalization and Recommender Systems in E-Commerce. 2007 AAAI; 2007

The main objective of this paper consists of improving recommendation effectiveness. An effective recommendation is one made to a user who is satisfied with the recommendation and motivated to buy the product in the future. The effectiveness of the recommendations depends on the information available for the filtering methods to predict if a product will please a user or not. In order to obtain this objective more specific objectives have been defined: Define a methodology to select the best structured source of information for the recommender systems. They must be defined criteria that allow sources to be qualified or evaluated to obtain the most relevant and reliable. Propose a mechanism to retrieve information about user preferences available on the Internet.