When Search Meets Recommendation: Learning Disentangled Search Representation for Recommendation
This addresses the challenge for online service providers like shopping platforms to better leverage combined search and recommendation data, though it is incremental as it builds on existing sequential recommendation methods with a novel disentanglement approach.
The paper tackles the problem of integrating user behavior data from both search and recommendation services, which have distinct intents, by proposing SESRec, a framework that disentangles similar and dissimilar representations from these behaviors to enhance recommendations, achieving consistent outperformance over state-of-the-art models in experiments on industrial and public datasets.
Modern online service providers such as online shopping platforms often provide both search and recommendation (S&R) services to meet different user needs. Rarely has there been any effective means of incorporating user behavior data from both S&R services. Most existing approaches either simply treat S&R behaviors separately, or jointly optimize them by aggregating data from both services, ignoring the fact that user intents in S&R can be distinctively different. In our paper, we propose a Search-Enhanced framework for the Sequential Recommendation (SESRec) that leverages users' search interests for recommendation, by disentangling similar and dissimilar representations within S&R behaviors. Specifically, SESRec first aligns query and item embeddings based on users' query-item interactions for the computations of their similarities. Two transformer encoders are used to learn the contextual representations of S&R behaviors independently. Then a contrastive learning task is designed to supervise the disentanglement of similar and dissimilar representations from behavior sequences of S&R. Finally, we extract user interests by the attention mechanism from three perspectives, i.e., the contextual representations, the two separated behaviors containing similar and dissimilar interests. Extensive experiments on both industrial and public datasets demonstrate that SESRec consistently outperforms state-of-the-art models. Empirical studies further validate that SESRec successfully disentangle similar and dissimilar user interests from their S&R behaviors.