Intention enhanced mixed attentive model for session-based recommendation

Abstract

Session-based recommendation aims to generate recommendations for the next item of users’ interest based on a given session. In this manuscript, we develop intention enhanced mixed attentive model (IMAM) to generate session-based recommendations using two important factors: temporal patterns and estimates of users’ intentions. Unlike existing methods which primarily leverage complicated gated recurrent units to model the temporal patterns, IMAM models the temporal patterns using a light-weight while effective position-sensitive attention mechanism. In IMAM, we also leverage the estimate of users’ prospective preferences to signify important items, and generate better recommendations. Our experimental results demonstrate that IMAM models significantly outperform the state-of-the-art methods in six benchmark datasets, with an improvement as much as 19.2%. In addition, our run-time performance comparison demonstrates that during testing, IMAM models are much more efficient than the best baseline method, with a significant average speedup of 47.7 folds.