Deeply Fusing Reviews and Contents for Cold Start Users in Cross-Domain Recommendation Systems

Fu, Wenjing; Peng, Zhaohui; Wang, Senzhang; Xu, Yang; Li, Jin

doi:10.1609/aaai.v33i01.330194

Cited by 130 publications

(72 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cross-domain recommendation (CDR) [5,12,13,16,17,28], which aims to improve the recommendation performance by means of transferring information from the auxiliary domain to the target domain, is one of the promising ways to solve data sparsity and cold start problem. Generally, CDR can be categorized into two categories.…”

Section: Cross-domain Recommendationmentioning

confidence: 99%

“…Since cold start users do not have any interactions in target domain. The other one aims at infering the preferences of cold start users based on their preferences observed in other domains [5,12,17]. These methods assume that there exists overlap in information between users and/or items across different domains, and train a mapping function from the source-domain into the target-domain.…”

Section: Cross-domain Recommendationmentioning

confidence: 99%

“…Cross-domain recommendation (CDR) [5,12,16,17], which aims to improve the recommendation performance by means of transferring information from the source domain to the target domain, is one of the promising ways to solve data sparsity and cold start problem. These methods assume that there exists overlap in information between users and/or items across different domains, and train a mapping function from the source-domain into the target-domain.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Heterogeneous Information Network based Cross Domain Insurance Recommendation System for Cold Start Users

Bi¹,

Song²,

Yao³

et al. 2020

Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

Internet is changing the world, adapting to the trend of internet sales will bring revenue to traditional insurance companies. Online insurance is still in its early stages of development, where cold start problem (prospective customer) is one of the greatest challenges. In traditional e-commerce field, several cross-domain recommendation (CDR) methods have been studied to infer preferences of cold start users based on their preferences in other domains. However, these CDR methods couldn't be applied to insurance domain directly due to the domain's specific properties. In this paper, we propose a novel framework called a Heterogeneous information network based Cross Domain Insurance Recommendation (HCDIR) system for cold start users. Specifically, we first try to learn more effective user and item latent features in both source and target domains. In source domain, we employ gated recurrent unit (GRU) to module users' dynamic interests. In target domain, given the complexity of insurance products and the data sparsity problem, we construct an insurance heterogeneous information network (IHIN) based on data from PingAn Jinguanjia, the IHIN connects users, agents, insurance products and insurance product properties together, giving us richer information. Then we employ three-level (relational, node, and semantic) attention aggregations to get user and insurance product representations. After obtaining latent features of overlapping users, a feature mapping between the two domains is learned by multi-layer perceptron (MLP). We apply HCDIR on Jinguanjia dataset, and show HCDIR significantly outperforms the state-of-the-art solutions. CCS CONCEPTS • Applied computing → Online insurance; • Information systems → Data mining; • Networks → Data path algorithms.

show abstract

Section: Cross-domain Recommendationmentioning

confidence: 99%

Section: Cross-domain Recommendationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Heterogeneous Information Network based Cross Domain Insurance Recommendation System for Cold Start Users

Bi¹,

Song²,

Yao³

et al. 2020

Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval

View full text Add to dashboard Cite

show abstract

“…For example, to solve the ''out-of-vocabulary'' problem of cross-domain transfer, Cen et al [29] extract a new CNN based model that leverages both word-level and character-based representations. RC-DFM [30] uses SDAE (stacked denoising autoencoder) to combine reviews, content information, and ratings to make crossdomain recommendations, and uses source domain reviews and product content information to alleviate cold-start issues. But they have a common problem that all the review information is migrated from the source domain without filtering.…”

Section: Cross Domain Recommender Systemmentioning

confidence: 99%

RACRec: Review Aware Cross-Domain Recommendation for Fully-Cold-Start User

et al. 2020

View full text Add to dashboard Cite

Traditional recommendation algorithms such as matrix factorization, collaborative filtering perform poorly when lack of interactive information of user and product, known as the user cold-start problem, which may cut down the revenue of E-Commerce platform. Moreover, it is more challenging to generate recommendation lists for users who have no information at all because there is no preference information about them that could be leveraged, which is the user fully-cold-start problem. In this paper, a review aware cross-domain recommendation algorithm, called RACRec, is proposed to address the fullycold-start problem in the field of product recommendation. Firstly, reviews are dynamically selected by using the adjacency matrix. Secondly, domain-specific preference vectors and domain-shared preference vectors of the cold start user are extracted by a migration model. On the other hand, the product feature vector in the target domain, which is generated from review texts by encoder and decoder, is combined with preference vectors of the cold-start user to make the rating prediction. Experiments on the Amazon datasets reveal that RACRec outperforms the state-of-the-art recommendation algorithms.

show abstract

“…The main idea of these methods is to replace the role of ID embedding with different kinds of side-information. [7,31,34,40,44] use user features to handle the cold-start situation. [12,46] use relational information to avoid cold-start problem.…”

Section: Introductionmentioning

confidence: 99%

Task-distribution-aware Meta-learning for Cold-start CTR Prediction

Cao

Yang

et al. 2020

Proceedings of the 28th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Nowadays, click-through rate (CTR) prediction has achieved great success in online advertising. However, making desirable predictions for unseen ads is still challenging, which is known as the cold-start problem. To address such a problem in CTR prediction, meta-learning methods have recently emerged as a popular direction. In these approaches, the predictions for each user/item are regarded as individual tasks, then training a meta-learner on them to implement zero-shot/few-shot learning for unknown tasks. Though these approaches have effectively alleviated the cold-start problem, two facts are not paid enough attention, 1) the diversity of the task difficulty and 2) the perturbation of the task distribution. In this paper, we propose an adaptive loss that ensures the consistency between the task weight and difficulty. Interestingly, the loss function can also be viewed as a description of the worst-case performance under distribution perturbation. Moreover, we develop an algorithm, under the framework of gradient descent with maxoracle (GDmax), to minimize such an adaptive loss. Then we prove the algorithm can return to a stationary point of the adaptive loss. Finally, we implement our method on top of the meta-embedding framework and conduct experiments on three real-world datasets. The experiments show that our proposed method significantly improves the predictions in the cold-start scenario.

show abstract

Deeply Fusing Reviews and Contents for Cold Start Users in Cross-Domain Recommendation Systems

Cited by 130 publications

References 15 publications

A Heterogeneous Information Network based Cross Domain Insurance Recommendation System for Cold Start Users

A Heterogeneous Information Network based Cross Domain Insurance Recommendation System for Cold Start Users

RACRec: Review Aware Cross-Domain Recommendation for Fully-Cold-Start User

Task-distribution-aware Meta-learning for Cold-start CTR Prediction

Contact Info

Product

Resources

About