Recommendation in Heterogeneous Information Networks Based on Generalized Random Walk Model and Bayesian Personalized Ranking

Jiang, Zhengshen; Liu, Hongzhi; Fu, Bin; Wu, Zhonghai; Zou, Tao

doi:10.1145/3159652.3159715

Cited by 66 publications

(33 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Embed. 4 [17] 2017 CVAE [61] 2017 entity2rec [77] 2017 NFM [38] 2017 MFM [66] 2017 Focused FM [7] 2017 GB-CENT [121] 2017 CML [43] 2017 ATRank [123] 2018 Div-HeteRec [73] 2018 HeteLearn [48] 2018 RNNLatentCross [8] 2018 DDL [119] 2018…”

Section: Summary Of Related Workmentioning

confidence: 99%

“…We notice several relevant works that perform low-rank factorization or representation learning in heterogeneous graphs, such as [48,59,73,77,80,116,122]. e interactions of users and items can be represented by a heterogeneous graph of two node types.…”

Section: Heterogeneous Graphsmentioning

confidence: 99%

“…With di erent de nitions of f θ and p, the two-stage process may run only once or iteratively until convergence. e de nition of s(u, i) in surveyed papers includes PageRank [48,59], PathSim [116] and so on. e likelihood function p guides the similarity-related parameters θ to t the distribution objective of observed similarities s or ratings R. e objective may be given a ributes X as learning auxiliary.…”

Section: Heterogeneous Graphsmentioning

confidence: 99%

See 2 more Smart Citations

Attribute-Aware Recommender System Based on Collaborative Filtering: Survey and Classification

et al. 2020

View full text Add to dashboard Cite

A ribute-aware CF models aims at rating prediction given not only the historical rating from users to items, but also the information associated with users (e.g. age), items (e.g. price), or even ratings (e.g. rating time). is paper surveys works in the past decade developing a ribute-aware CF systems, and discovered that mathematically they can be classi ed into four di erent categories. We provide the readers not only the high level mathematical interpretation of the existing works in this area but also the mathematical insight for each category of models. Finally we provide in-depth experiment results comparing the e ectiveness of the major works in each category.

show abstract

Section: Summary Of Related Workmentioning

confidence: 99%

Section: Heterogeneous Graphsmentioning

confidence: 99%

Section: Heterogeneous Graphsmentioning

confidence: 99%

See 1 more Smart Citation

Attribute-Aware Recommender System Based on Collaborative Filtering: Survey and Classification

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Most HIN based recommendation models use meta-paths [Sun et al, 2011] to mine the semantic relations between users and items. As different metapaths can capture different semantics [Jiang et al, 2018;Liu et al, 2019] and result in different recommendation lists [Shi et al, 2015], information extracted from different metapaths should be fused properly.…”

Section: Introductionmentioning

confidence: 99%

“…Compared with traditional recommendation model like matrix factorization [Koren et al, 2009], HIN based methods have achieved performance improvement to some extent. However, there still exist two problems for these models.…”

Section: Introductionmentioning

confidence: 99%

Unified Embedding Model over Heterogeneous Information Network for Personalized Recommendation

Wang

Liu

et al. 2019

Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence

Self Cite

View full text Add to dashboard Cite

Most of heterogeneous information network (HIN) based recommendation models are based on the user and item modeling with meta-paths. However, they always model users and items in isolation under each meta-path, which may lead to information extraction misled. In addition, they only consider structural features of HINs when modeling users and items during exploring HINs, which may lead to useful information for recommendation lost irreversibly. To address these problems, we propose a HIN based unified embedding model for recommendation, called HueRec. We assume there exist some common characteristics under different meta-paths for each user or item, and use data from all meta-paths to learn unified users’ and items’ representations. So the interrelation between meta-paths are utilized to alleviate the problems of data sparsity and noises on one meta-path. Different from existing models which first explore HINs then make recommendations, we combine these two parts into an end-to-end model to avoid useful information lost in initial phases. In addition, we embed all users, items and meta-paths into related latent spaces. Therefore, we can measure users’ preferences on meta-paths to improve the performances of personalized recommendation. Extensive experiments show HueRec consistently outperforms state-of-the-art methods.

show abstract

Semantic Enhanced Top-k Similarity Search on Heterogeneous Information Networks

Zhang

et al. 2020

Database Systems for Advanced Applications

View full text Add to dashboard Cite

Recommendation in Heterogeneous Information Networks Based on Generalized Random Walk Model and Bayesian Personalized Ranking

Cited by 66 publications

References 17 publications

Attribute-Aware Recommender System Based on Collaborative Filtering: Survey and Classification

Attribute-Aware Recommender System Based on Collaborative Filtering: Survey and Classification

Unified Embedding Model over Heterogeneous Information Network for Personalized Recommendation

Semantic Enhanced Top-k Similarity Search on Heterogeneous Information Networks

Contact Info

Product

Resources

About