Personalized Case-Based Explanation of Matrix Factorization Recommendations

• Published in: Case-Based Reasoning Research and Development Vol. 11680; pp. 140 - 154
• Main Authors: Jorro-Aragoneses, Jose, Caro-Martinez, Marta, Recio-Garcia, Juan Antonio, Diaz-Agudo, Belen, Jimenez-Diaz, Guillermo
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer International Publishing AG 2019; Springer International Publishing
• Series: Lecture Notes in Computer Science
• Subjects: Case-based explanation; Matrix factorization; Personalised explanation
• ISBN: 3030292487; 9783030292485
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-29249-2_10

Matrix factorization is an advanced recommendation strategy based on characterizing both items and users on a vector of latent factors inferred from rating patterns. These vectors represent, somehow, a characterization of the user preferences in a lower dimensionality space. Although matrix factorization is more accurate that other recommendation strategies, the main problem associated with this approach is that the discovered factors are opaque and difficult to explain to the final user. In this paper we propose a personalized case-based explanation strategy that uses the latent factors to find similar explanatory cases already rated by the user.