Identifying Information from heterogeneous and distributed information sources for recommender systems

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

Mexico

Conferencia; Fourth Mexican International Conference on Artificial Intelligence (MICAI 2005); 2005

With easy access to World Wide Web the users are overloaded of information. Recommender systems have emerged as research approach to address this problem. The users want to find what they need, when they need it and under the conditions that they want. These conditions drive to the recommender systems to access to different sources to find relevant information to recommend. In this paper we presented a set of intrinsic characteristics to determine the relevance of the sources to make recommendations. These characteristics allow to have a representation of the information contained in the sources that is relevant to recommend and a set of criteria to select the most relevant. A multi-agent system has been designed to obtain these characteristics. Preliminary results of recommendations made with the selected sources are presented in this paper.