Symbolic data analysis tools for recommendation systems

Recommendation Systems have become an important tool to cope with the information overload problem by acquiring data about the user behavior. After tracing the user behavior, through actions or rates, Computational Recommendation Systems use information filtering techniques to recommend items. In order to recommend new items, one of the three approaches has been mainly adopted: Content Based Filtering, Collaborative Filtering or hybrid filtering methods. This paper presents three information filtering methods, each of them based on one of the previous approaches. In our methods, the user profile is designed through Symbolic Data Structures and the user and item correlations are computed through distance functions adapted from the Symbolic Data Analysis domain. The usage of Symbolic Data Analysis tools have improved the performance of Recommendation Systems, specially when there is little information about the user.

• Type: Journal
• Year: 2011
• Where: KAIS
• Authors: Byron Leite Dantas Bezerra, Francisco de Assis Tenorio de Carvalho
• Keywords: KAIS 2011, Recommendation Systems, Symbolic Data, Symbolic Data Analysis