Complementary Product Recommendation for Long-tail Products

Identifying complementary relations between products plays a key role in e-commerce Recommender Systems (RS). Existing methods  in Complementary Product Recommendation (CPR), however, focus only on identifying complementary relations in huge and data-rich  catalogs, while none of them considers real-world scenarios of small and medium e-commerce platforms with limited number of  interactions. In this paper, we discuss our research proposal that addresses the problem of identifying complementary relations in  such sparse settings. To overcome the data sparsity problem, we propose to first learn complementary relations in large and data-rich  catalogs and then transfer learned knowledge to small and scarce ones. To be able to map individual products across different catalogs  and thus transfer learned relations between them, we propose to create Product Universal Embedding Space (PUES) using textual and visual product meta-data, which serves as a common ground for the products from arbitrary catalog.

Cite: Rastislav Papso. 2023. Complementary Product Recommendation for Long-tail Products. In Proceedings of the 17th ACM Conference on Recommender Systems (RecSys ’23). Association for Computing Machinery, New York, NY, USA, 1305–1311. https://doi.org/10.1145/3604915.3608864