Joint Training Capsule Network for Cold Start Recommendation
This addresses the problem of recommending items to new users with limited interaction history, representing an incremental advance in recommendation systems.
The paper tackles the cold start recommendation problem by proposing a joint training capsule network (JTCN) that mimics high-level user preferences from side information, achieving improvements of at least 7.07% on CiteULike and 16.85% on Amazon in Recall@100.
This paper proposes a novel neural network, joint training capsule network (JTCN), for the cold start recommendation task. We propose to mimic the high-level user preference other than the raw interaction history based on the side information for the fresh users. Specifically, an attentive capsule layer is proposed to aggregate high-level user preference from the low-level interaction history via a dynamic routing-by-agreement mechanism. Moreover, JTCN jointly trains the loss for mimicking the user preference and the softmax loss for the recommendation together in an end-to-end manner. Experiments on two publicly available datasets demonstrate the effectiveness of the proposed model. JTCN improves other state-of-the-art methods at least 7.07% for CiteULike and 16.85% for Amazon in terms of Recall@100 in cold start recommendation.