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

Liu, Weiming; Zheng, Xiaolin; Hu, Mengling; Chen, Chaochao

doi:10.1145/3485447.3512166

Cited by 40 publications

(11 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Transfer-based methods [19,42] learn a linear or nonlinear mapping function across domains. Multitask-based methods [9,16,47] enable dual knowledge transfer by introducing shared connection modules in neural networks. Clustered-based methods [35] adopt co-clustering approach to learn cross-domain comprehensive embeddings by collectively leveraging single-domain and cross-domain information within a unified framework.…”

Section: Related Workmentioning

confidence: 99%

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

Zheng

Liu

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

Self Cite

View full text Add to dashboard Cite

Sequential Recommendation (SR) characterizes evolving patterns of user behaviors by modeling how users transit among items. However, the short interaction sequences limit the performance of existing SR. To solve this problem, we focus on Cross-Domain Sequential Recommendation (CDSR) in this paper, which aims to leverage information from other domains to improve the sequential recommendation performance of a single domain. Solving CDSR is challenging. On the one hand, how to retain single domain preferences as well as integrate cross-domain influence remains an essential problem. On the other hand, the data sparsity problem cannot be totally solved by simply utilizing knowledge from other domains, due to the limited length of the merged sequences. To address the challenges, we propose DDGHM, a novel framework for the CDSR problem, which includes two main modules, i.e., dual dynamic graph modeling and hybrid metric training. The former captures intra-domain and inter-domain sequential transitions through dynamically constructing two-level graphs, i.e., the local graphs and the global graphs, and incorporating them with a fuse attentive gating mechanism. The latter enhances user and item representations by employing hybrid metric learning, including collaborative metric for achieving alignment and contrastive metric for preserving uniformity, to further alleviate data sparsity issue and improve prediction accuracy. We conduct experiments on two benchmark datasets and the results demonstrate the effectiveness of DDGHM.

show abstract

Section: Related Workmentioning

confidence: 99%

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

Zheng

Liu

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

Self Cite

View full text Add to dashboard Cite

show abstract

“…As a result, the trained layers or embedding can learn the knowledge from both domains, and hence can provide more accurate recommendations than single-domain recommendations. In contrast, transfer learning-based CDRSs focus on recommendations in the target domain [16]- [18]. These CDRSs extract knowledge from the source domain and use the learned knowledge to improve the target recommendations.…”

Section: A Motivationmentioning

confidence: 99%

“…In addition, user matching is an arduous task and may involve privacy issues, particu-larly when data come from different companies. Some CDRSs, such as MMT-DRR [20], RecSys-DAN [21], and CFAA [18], therefore extract user interaction patterns from the source domain and transfer the learned interaction patterns to the target domain, where the interaction patterns are defined as learned user (item) embedding or the distribution of predictions. They define domains as different categories in a Web service, such as Amazon Book and Amazon Movie, or different places in a real-world service, such as the restaurant visit records in different cities.…”

Section: A Motivationmentioning

confidence: 99%

“…However, MMT-DRR cannot work without domain-shared contextual information. ESAM [31] and CFAA [18] removed the requirement of domain-shared contextual information and aligned the attribution distribution and correlation between source and target embedding spaces to transfer knowledge. Besides, RecSys-DAN [21] proposed a novel discriminator and minimized the divergence of the predictions between the source domain and the target domain for knowledge transformation.…”

Section: Related Work a Cross-domain Recommender Systemsmentioning

confidence: 99%

“…• CFAA [18] proposes an embedding attribution alignment module to reduce the discrepancy of attribution distributions and relations between source and target domains.…”

Section: B) Cross-domain Recommendations (Cdrs)mentioning

confidence: 99%

See 2 more Smart Citations

Debiasing Graph Transfer Learning via Item Semantic Clustering for Cross-Domain Recommendations

Li¹,

Amagata²,

Zhang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep learning-based recommender systems may lead to over-fitting when lacking training interaction data. This over-fitting significantly degrades recommendation performances. To address this data sparsity problem, cross-domain recommender systems (CDRSs) exploit the data from an auxiliary source domain to facilitate the recommendation on the sparse target domain. Most existing CDRSs rely on overlapping users or items to connect domains and transfer knowledge. However, matching users is an arduous task and may involve privacy issues when data comes from different companies, resulting in a limited application for the above CDRSs. Some studies develop CDRSs that require no overlapping users and items by transferring learned user interaction patterns. However, they ignore the bias in user interaction patterns between domains and hence suffer from an inferior performance compared with single-domain recommender systems. In this paper, based on the above findings, we propose a novel CDRS, namely semantic clustering enhanced debiasing graph neural recommender system (SCDGN), that requires no overlapping users and items and can handle the domain bias. More precisely, SCDGN semantically clusters items from both domains and constructs a cross-domain bipartite graph generated from item clusters and users. Then, the knowledge is transferred via this cross-domain user-cluster graph from source to the target. Furthermore, we design a debiasing graph convolutional layer for SCDGN to extract unbiased structural knowledge from the cross-domain user-cluster graph. Our Experimental results on three public datasets and a pair of proprietary datasets verify the effectiveness of SCDGN over stateof-the-art models in terms of cross-domain recommendations.

show abstract

Semantic Relation Transfer for Non-overlapped Cross-domain Recommendations

Amagata

Zhang

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

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

Cited by 40 publications

References 30 publications

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

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

Debiasing Graph Transfer Learning via Item Semantic Clustering for Cross-Domain Recommendations

Semantic Relation Transfer for Non-overlapped Cross-domain Recommendations

Contact Info

Product

Resources

About