FFM-SVD: A Novel Approach for Personality-aware Recommender Systems

Abstract

This paper addresses and evaluates approaches to incorporating personality data into a recommender system. Automatic personality recognition is enabled by the LIWC dictionary. Personality-aware pre-filtering techniques are developed and discussed, with the introduced non-targeted stratified personality sampling performing the best. A novel personality-aware model, FFM-SVD, is proposed and shown to outperform alternative models in prediction accuracy.