DCDIR: A Deep Cross-Domain Recommendation System for Cold Start Users in Insurance Domain

Ye Bi

[0]

Liqiang Song

Mengqiu Yao

Zhenyu Wu

Jianming Wang

Jing Xiao

SIGIR '20: The 43rd International ACM SIGIR conference on research and development in Information Retrieval Virtual Event China July, 2020, pp. 1661-1664, 2020.

Cited by: 0|Bibtex|Views35|DOI:https://doi.org/10.1145/3397271.3401193

Other Links: dl.acm.org|arxiv.org|dblp.uni-trier.de|academic.microsoft.com

Keywords:

source domainmulti layer perceptioncollaborate filteringitem latent featureKnowledge GraphMore(24+)

Weibo:

To deal with insurance product complexity and cold start problem, we propose Deep Cross Domain Insurance Recommendation System for cold start users

Abstract:

Internet insurance products are apparently different from traditional e-commerce goods for their complexity, low purchasing frequency, etc. So, cold start problem is even worse. In traditional e-commerce field, several cross-domain recommendation (CDR) methods have been studied to infer preferences of cold start users based on their prefe...More

Code:

Data:

Insight

Full Text

PPT (Upload PPT)

Introduction

Internet finance is booming and rapidly infiltrating into all kinds of traditional financial fields.
DCDIR: A Deep Cross-Domain Recommendation System for Cold Start Users in Insurance Domain.
The authors propose a novel framework called a Deep Cross-Domain Insurance Recommendation System (DCDIR) for cold start users.

Highlights

Nowadays, internet finance is booming and rapidly infiltrating into all kinds of traditional financial fields
We propose a novel framework called a Deep Cross-Domain Insurance Recommendation System (DCDIR) for cold start users
To provide recommendations to cold start users, we propose Deep Cross Domain Insurance Recommendation System (DCDIR)
We conduct extensive experiments to answer the following questions: RQ1: How does DCDIR model perform compared with baselines in terms of NDCG and Recall@3? RQ2: Can DCDIR alleviate the data sparsity problem? RQ3: How does path-based insurance knowledge graph (ISKG) module affect the performance of DCDIR for cold start users?
Overall, benefiting from the proposed insurance products’ KG path-based representations and source domain information, DCDIR beats all comparative methods, and achieves the range of 0.22%-8.59% and 0.51%-26.23% improvements over the best comparative model in Recall@3 and NDCG under all levels of data sparsity, respectively. These experiments reveal a number of interesting discoveries: (1) All cross-domain methods yield better performances than single-domain methods with mixture of target and source domain data, demonstrating the importance of cross-domain module; (2) Owing to the capability of using insurance productsâĂŹ knowledge, three variants of DCDIR (DCDIR, DCDIR-V1 and DCDIR-V2) defeat other comparative methods; (3) It demonstrates that DCDIR achieves more improvements in a sparser dataset than in a denser one
To deal with insurance product complexity and cold start problem, we propose DCDIR for cold start users

Results

The authors first try to learn more effective user and item latent features in both source and target domains.
Given the complexity of insurance products, the authors design a meta-path based method over the knowledge graph the authors constructed.
For the complexity of insurance products, the authors design a metapath based method to learn more effective latent user and item features, revealing reasons behind recommendations.
Given rating matrices and ISKG, the goal is to learn the mapping function from nonfinancial domain to insurance domain, which can help them deal with cold start users.
To help users better understand insurance products, the authors design a meta-path based method.
RQ3: How does path-based ISKG module affect the performance of DCDIR for cold start users?
The authors randomly select 30% of the total overlapped users and remove their information in the target domain as cold start users for evaluating the performance.
Overall, benefiting from the proposed insurance products’ KG path-based representations and source domain information, DCDIR beats all comparative methods, and achieves the range of 0.22%-8.59% and 0.51%-26.23% improvements over the best comparative model in Recall@3 and NDCG under all levels of data sparsity, respectively.
These experiments reveal a number of interesting discoveries: (1) All cross-domain methods yield better performances than single-domain methods with mixture of target and source domain data , demonstrating the importance of cross-domain module; (2) Owing to the capability of using insurance productsâĂŹ knowledge, three variants of DCDIR (DCDIR, DCDIR-V1 and DCDIR-V2) defeat other comparative methods; (3) It demonstrates that DCDIR achieves more improvements in a sparser dataset than in a denser one.
4.3 The impact of meta-path based ISKG module to cold start users (RQ 3)
It is necessary to study if the designed meta-path based ISKG module can deal with cold start users problem in an effective way.

Conclusion

Table 3 indicates that, suffering from the cold start problem, DCDIR’s best parameters in ISKG module are path number as 20 and choosing path strategy is the designed top K method in terms of Recall@3 and NCDG.
Given the complexity of insurance products, the authors design a meta-path based method over insurance product knowledge graph, which can provide interpretable recommendations to users.
The authors apply DCDIR on the companyâĂŹs dataset, and show DCDIR significantly outperforms the state-of-the-art solutions

Summary

Internet finance is booming and rapidly infiltrating into all kinds of traditional financial fields.
DCDIR: A Deep Cross-Domain Recommendation System for Cold Start Users in Insurance Domain.
The authors propose a novel framework called a Deep Cross-Domain Insurance Recommendation System (DCDIR) for cold start users.
The authors first try to learn more effective user and item latent features in both source and target domains.
Given the complexity of insurance products, the authors design a meta-path based method over the knowledge graph the authors constructed.
For the complexity of insurance products, the authors design a metapath based method to learn more effective latent user and item features, revealing reasons behind recommendations.
Given rating matrices and ISKG, the goal is to learn the mapping function from nonfinancial domain to insurance domain, which can help them deal with cold start users.
To help users better understand insurance products, the authors design a meta-path based method.
RQ3: How does path-based ISKG module affect the performance of DCDIR for cold start users?
The authors randomly select 30% of the total overlapped users and remove their information in the target domain as cold start users for evaluating the performance.
Overall, benefiting from the proposed insurance products’ KG path-based representations and source domain information, DCDIR beats all comparative methods, and achieves the range of 0.22%-8.59% and 0.51%-26.23% improvements over the best comparative model in Recall@3 and NDCG under all levels of data sparsity, respectively.
These experiments reveal a number of interesting discoveries: (1) All cross-domain methods yield better performances than single-domain methods with mixture of target and source domain data , demonstrating the importance of cross-domain module; (2) Owing to the capability of using insurance productsâĂŹ knowledge, three variants of DCDIR (DCDIR, DCDIR-V1 and DCDIR-V2) defeat other comparative methods; (3) It demonstrates that DCDIR achieves more improvements in a sparser dataset than in a denser one.
4.3 The impact of meta-path based ISKG module to cold start users (RQ 3)
It is necessary to study if the designed meta-path based ISKG module can deal with cold start users problem in an effective way.
Table 3 indicates that, suffering from the cold start problem, DCDIR’s best parameters in ISKG module are path number as 20 and choosing path strategy is the designed top K method in terms of Recall@3 and NCDG.
Given the complexity of insurance products, the authors design a meta-path based method over insurance product knowledge graph, which can provide interpretable recommendations to users.
The authors apply DCDIR on the companyâĂŹs dataset, and show DCDIR significantly outperforms the state-of-the-art solutions

Tables

Table1: Statistics of the JGJISNF dataset
Table2: Performance comparison in Recall@3 and NDCG. The best baseline except DCDIR is bolded. Numbers in “()” represent the percentage of three variants’ performance at η=10% compared with their best performance in other sparsity level
Table3: Performance comparison in Recall@3 and NDCG under a sparse setting (η=10%) with changing path number and choosing path strategy

Download tables as Excel

Reference

X. He, L. Liao, H. Zhang, L. Nie, X. Hu, and T. Chua. 2017. Neural Collaborative Filtering. In WWW. 173–182.
B. Hidasi, A. Karatzoglou, L. Baltrunas, and D. Tikk. 2016. Session-based Recommendations with Recurrent Neural Networks. In ICLR.
G. Ji, S. He, L. Xu, K. Liu, and J. Zhao. 2015. Knowledge Graph Embedding via Dynamic Mapping Matrix. In ACL. 687–696.
S. K. Kang, J. Hwang, D. Lee, and H. Yu. 2019. Semi-Supervised Learning for Cross-Domain Recommendation to Cold-Start Users. In CIKM. 1563–1572.
Z. Liu, C. Zang, K. Kuang, H. Zou, H. Zheng, and P. Cui. 2019. Causation-Driven Visualizations for Insurance Recommendation. In ICME Workshops. 471–476.
M. Ma, P. Ren, Y. Lin, Z. Chen, J. Ma, and M. D. Rijke. 2019. π -Net: A Parallel Information-sharing Network for Shared-account Cross-domain Sequential Recommendations. In SIGIR. 685–694.
T. Man, H. Shen, X. Jin, and X. Cheng. 201Cross-Domain Recommendation: An Embedding and Mapping Approach. In IJCAI. 2464–2470.
T. Mikolov, I. Sutskever, K. Chen, G. S. Corrado, and J. Dean. 2013. Distributed Representations of Words and Phrases and their Compositionality. In NIPS.
M. Qazi, G. M. Fung, K. J. Meissner, and E. R. Fontes. 2017. An Insurance Recommendation System Using Bayesian Networks. In RecSys. 274–278.
S. Rendle, C. Freudenthaler, Z. Gantner, and L. S. Thieme. 2009. BPR: Bayesian Personalized Ranking from Implicit Feedback. In UAI. 452–461.
L. Rokach, G. Shani, B. Shapira, E. Chapnik, and G. Siboni. 2013. Recommending insurance riders. In SAC. 253–260.
Y. Sun, J. Han, X. Yan, P. S. Yu, and T. Wu. 2011. PathSim: Meta Path-Based Top-K Similarity Search in Heterogeneous Information Networks. In PVLDB.
X. Wang, X. He, Y. Cao, M. Liu, and T. Chua. 2019. KGAT: Knowledge Graph Attention Network for Recommendation. In SIGKDD. 950–958.
X. Wang, D. Wang, C. Xu, X. He, Y. Cao, and T. Chua. 2019. Explainable Reasoning over Knowledge Graphs for Recommendation. In AAAI. 5329–5336.

Full Text

Your rating :