Abstract:Learning accurate cross -domain preference mappings in the absence of overlapped users/items has presented a persistent challenge in Non-overlapping Cross-domain Recommendation (NOCDR). Despite the efforts made in previous studies to address NOCDR, several limitations still exist. Specifically, 1) while some approaches substitute overlapping users/items with overlapping behaviors, they cannot handle NOCDR scenarios where such auxiliary information is unavailable; 2) often, cross domain preference mapping is modeled by learning deterministic explicit representation matchings between sampled users in two domains. However, this can be biased due to individual preferences and thus fails to incorporate preference continuity and universality of the general population. In light of this, we assume that despite the scattered nature of user behaviors, there exists a consistent latent preference distribution shared among common people. Modeling such distributions further allows us to capture the continuity in user behaviors within each domain and discover preference invariance across domains. To this end, we propose a Distributional domain-invariant Preference Matching method for non-overlapping Cross-Domain Recommendation (DPMCDR). For each domain, we hierarchically approximate a posterior of domain-level preference distribution with empirical evidence derived from user-item interactions. Next, we aim to build distributional implicit matchings between the domain-level preferences of two domains. This process involves mapping them to a shared latent space and seeking a consensus on domain invariant preference by minimizing the distance between their distributional representations therein. In this way, we can identify the alignment of two non-overlapping domains if they exhibit similar patterns of domain-invariant preference. Experiments on real-world datasets demonstrate that DPMCDR outperforms the state-of-the-art approaches with a range of evaluation metrics.

User Distribution Mapping Modelling with Collaborative Filtering for Cross Domain Recommendation

Cross-Domain Recommendation for Cold-Start Users via Neighborhood Based Feature Mapping

Modeling Domains As Distributions with Uncertainty for Cross-Domain Recommendation

Cross-Domain Recommendation Via User-Clustering and Multidimensional Information Fusion

Mining User Consistent and Robust Preference for Unified Cross Domain Recommendation

DTCDR: A Framework for Dual-Target Cross-Domain Recommendation

Low-dimensional Alignment for Cross-Domain Recommendation.

Diffusion Cross-domain Recommendation

Distributional Domain-Invariant Preference Matching for Cross-Domain Recommendation

Cross-Domain Recommendation Based on Sentiment Analysis and Latent Feature Mapping

DDGHM: Dual Dynamic Graph with Hybrid Metric Training for Cross-Domain Sequential Recommendation

DDTCDR: Deep Dual Transfer Cross Domain Recommendation

Collaborative Filtering with Attribution Alignment for Review-based Non-overlapped Cross Domain Recommendation

Leveraging Distribution Alignment Via Stein Path for Cross-Domain Cold-Start Recommendation

Heterogeneous and Clustering-Enhanced Personalized Preference Transfer for Cross-Domain Recommendation

Dual Metric Learning for Effective and Efficient Cross-Domain Recommendations

Transfer-Meta Framework for Cross-domain Recommendation to Cold-Start Users

Self-Supervised Cross Domain Social Recommendation

Cross-Domain Recommendation: An Embedding and Mapping Approach

Collaborative Filtering Recommendation Based on Multi-Domain Semantic Fusion

Exploiting Variational Domain-Invariant User Embedding for Partially Overlapped Cross Domain Recommendation