Cross domain recommendation improves next item prediction with user transfer
Cross-Domain Sequential Recommendation with Representation-Level User Transfer
Information RetrievalArtificial IntelligenceDistributed, Parallel, and Cluster Computing
Summary
This paper looks at how to make better recommendations for what someone might want next, like in shopping or streaming, by using their behavior from different places without sharing private data. The authors created a method that moves learned user information between areas without sharing actual user actions or items. Their system works well even for users who appear in only one area, making the recommendation model flexible and practical. Tests showed their way gave better results in several realistic conditions compared to older methods.
What this means in practice
- •For e-commerce platform engineers: Integrate user behavior from multiple markets to improve personalized product recommendations without sharing raw data between domains.$Commercial implications: Enables privacy-preserving cross-market personalization improving recommendation engines for online retailers.
- •For music streaming developers: Enhance next-song recommendations by transferring user preferences learned from different genre domains without exposing sensitive user interaction logs.
Authors
Abdulaziz Samra, Nikita Sukhorukov, Alexey Grishanov, Alexey Vasilev, Evgeny Frolov
Abstract
Cross-domain sequential recommendation aims to improve next-item prediction in a target domain by leveraging user behavior from auxiliary source domains. However, real-world deployment faces two critical challenges: business constraints that isolate interaction data across domains, and user heterogeneity, where many users lack cross-domain activity. This makes existing cross-domain methods difficult to apply. In this paper, we propose a simple and practical framework for cross-domain sequential recommendations that transfers information exclusively through learned user representations, without sharing raw interaction data or item embeddings across domains. We introduce a lightweight adapter that projects the source-domain user representation into the target-domain embedding space and integrates it with the target interaction sequence. Importantly, when source data is unavailable (i.e., the user observed only in the target domain), the model seamlessly defaults to a robust target-only sequential recommender, retaining benefits from joint training across all users. Extensive experiments were conducted under two setups; one targets benchmarking against previous baselines, while the other focuses on realistic and scalable conditions. Both sets of experiments demonstrate consistent performance gains for shared and unshared users.