TOWARDS TEMPORAL ABSENCE MODELLING: TEMPORAL ABSENCE CONNOTATION IN NETFLIX PRIZE DATA

Research on evaluating recommender systems shows that algorithms in this area are still deficient in prediction accuracy but recent works prove that modeling with temporal dynamics improves the degree of recommendation accuracy. Recommendations are invariably based on similarities of users and/or items in the user-item matrix of a system, user profiles, and rating information which presumes the presence of users or items in the matrix. The major difference being in the way the algorithms analyze data sources to develop notions of affinity between users for use in identifying well matched pairs. Not many have focused on the temporal absence as an indicator of preference or concept drift: and hence a factor for inclusion in the recommender algorithms and models to improve accuracy. In this paper we to define temporal absence in the context of recommender systems and find out, through examination of the Netflix Prize data, the extent of temporal absence and the significance of such information in future research and improvement of recommendation algorithms.