When Search Meets Recommendation: Learning Disentangled Search Representation for Recommendation
Open Access
- 18 July 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1313-1323
Abstract
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.Keywords
All Related Versions
Funding Information
- Beijing Outstanding Young Scientist Program (BJJWZYJH012019100020098)
- National Key Research and Development Program of China (2019YFE0198200)
This publication has 26 references indexed in Scilit:
- Deep Interest Evolution Network for Click-Through Rate PredictionProceedings of the AAAI Conference on Artificial Intelligence, 2019
- Attentive Long Short-Term Preference Modeling for Personalized Product SearchACM Transactions on Information Systems, 2019
- Self-Attentive Sequential Recommendation2018
- Deep Interest Network for Click-Through Rate Prediction2018
- STAMP2018
- Improved Fusion of Visual and Language Representations by Dense Symmetric Co-attention for Visual Question Answering2018
- Personalized Top-N Sequential Recommendation via Convolutional Sequence Embedding2018
- Learning Latent Vector Spaces for Product Search2016
- Ups and Downs2016
- Divergence measures based on the Shannon entropyIEEE Transactions on Information Theory, 1991