Learning to Structure Long-term Dependence for Sequential Recommendation
This work addresses the underexplored problem of long-term dependence in sequential recommendation, which is incremental as it builds on existing methods by incorporating intent inference from side-information.
The paper tackles the challenge of modeling long-term dependence in sequential recommendation by proposing GatedLongRec, which extracts distant actions related to user intent using a top-k gating network and encodes transition patterns, resulting in significant improvements on two large datasets.
Sequential recommendation recommends items based on sequences of users' historical actions. The key challenge in it is how to effectively model the influence from distant actions to the action to be predicted, i.e., recognizing the long-term dependence structure; and it remains an underexplored problem. To better model the long-term dependence structure, we propose a GatedLongRec solution in this work. To account for the long-term dependence, GatedLongRec extracts distant actions of top-$k$ related categories to the user's ongoing intent with a top-$k$ gating network, and utilizes a long-term encoder to encode the transition patterns among these identified actions. As user intent is not directly observable, we take advantage of available side-information about the actions, i.e., the category of their associated items, to infer the intents. End-to-end training is performed to estimate the intent representation and predict the next action for sequential recommendation. Extensive experiments on two large datasets show that the proposed solution can recognize the structure of long-term dependence, thus greatly improving the sequential recommendation.