User Distribution Mapping Modelling with Collaborative Filtering for Cross Domain Recommendation

Weiming Liu, Chaochao Chen, Xinting Liao, Mengling Hu, Jiajie Su, Yanchao Tan, Fan Wang

Keywords: Recommendation

Abstract: User cold-start recommendation aims to provide accurate items for the newly joint users and is a hot and challenging problem. Nowadays as people participate in different domains, how to recommend items in the new domain for users in an old domain has become more urgent. In this paper, we focus on the Dual Cold-Start Cross Domain Recommendation (Dual-CSCDR) problem. That is, providing the most relevant items for new users on the source and target domains. The prime task in Dual-CSCDR is to properly model user-item rating interactions and map user expressive embeddings across domains. However, previous approaches cannot solve Dual-CSCDR well, since they separate the collaborative filtering and feature mapping process, leading to the error superimposition issue. Moreover, most of these methods fail to fully exploit the cross-domain relationship among a large number of non-overlapped users, which strongly limits their performance. To fill this gap, we propose the User Embedding Distribution Alignment model with Collaborative Filtering (UEDMCF), a novel end-to-end cold-start cross-domain recommendation framework for the Dual-CSCDR problem. UEDMCF includes two main modules, i.e., rating prediction module and feature space alignment module. The former module adopts one-hot ID vectors and multi-hot historical ratings for collaborative filtering via a contrastive loss. The latter module contains overlapped user embedding alignment and general user distribution alignment. Specifically, we innovatively propose unbalanced distribution optimal transport with a typical subgroup discovering algorithm to map the whole user distributions. Our empirical study on several datasets demonstrates that UEDMCF significantly outperforms the state-of-the-art models under the Dual-CSCDR setting.

Track: Graph Algorithms and Learning for the Web

Submission Guidelines Scope: Yes

Submission Guidelines Blind: Yes

Submission Guidelines Format: Yes

Submission Guidelines Limit: Yes

Submission Guidelines Authorship: Yes

Student Author: Yes

Submission Number: 119