Recommender systems with social regularization

Ma, Hao; Zhou, Dengyong; Liu, Chao; Lyu, Michael R.; King, Irwin

doi:10.1145/1935826.1935877

Cited by 1,380 publications

(948 citation statements)

References 32 publications

Supporting

Mentioning

937

Contrasting

Unclassified

Order By: Relevance

“…Keywords: recommender system, context-awareness, collaborative ranking, behavior modeling the scenarios with user explicit feedbacks [6,7], or cast the recommendation task as user adoption predictions in the scenarios without explicit feedbacks [4,9].…”

Section: Related Workmentioning

confidence: 99%

“…Social information is utilized to better shape the user latent space typically. For example, [6] makes a recommendation by adding additional social regularization terms in MF to constrain the user latent feature vectors to be similar to his or her friends' average latent features. [12] proposes a category-specific social trust influence weight which outlines several variants of weighting friends within circles based on their inferred expertise levels.…”

Section: Related Workmentioning

confidence: 99%

“…Where N is the length of recommendation list and We compare the proposed methods with baseline methods: Item-based CF [10], User-based CF, BMF, KDDCUP07 [4], SoReg [6], SoCo [5], CTR [2], and some reduction versions of UATM.…”

Section: Experimental Protocolmentioning

confidence: 99%

See 2 more Smart Citations

User Adoption Tendency Modeling for Social Contextual Recommendation

Wang

Liu

et al. 2015

MATEC Web of Conferences

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

User Adoption Tendency Modeling for Social Contextual Recommendation

Wang

Liu

et al. 2015

MATEC Web of Conferences

View full text Add to dashboard Cite

“…As such, the proximity measures by using random walks have clear physical meanings than other latent methods [27,33].…”

Section: Problem Formulationmentioning

confidence: 99%

A proximity measure for link prediction in social user-item networks

Chang

Lee

2014

Proceedings of the 2014 IEEE 15th International Conference on Information Reuse and Integration (IEEE IRI 2014)

View full text Add to dashboard Cite

“…(Soo, 2004) [11] Proposed a prototype design for building a personalized recommender system to recommend travel related information according to users' contextual information. (M.-H. Park, 2007) [9] Proposed a location based personalized recommender system, which can reflect users' personal preferences by modeling user contextual information through Bayesian Networks.(Hao Ma) [17] Analyse latent factor using probabilistic matrix factorization, we learn the user latent feature space and item latent feature space by employing a user social network and a user-item matrix simultaneously and seamlessly. (H. Ma, 2011) [22] Proposed a matrix factorization framework with social regularization.…”

mentioning

confidence: 99%

Scalable Recommendation With Social Contextual Information

.¹

2015

IJRET

View full text Add to dashboard Cite

Recommender systems are used to achieve effective and useful results in a social networks. The social recommendation will provide a social network structure but it is challenging to fuse social contextual factors which are derived from user ' Keywords: social recommendation, individual preference, interpersonal influence, matrix factorization.Social network users generate large volumes of information, which makes it necessary to exploit highly accurate recommender systems to assist them in finding useful results. Social contextual factors work on the contextual information and integrate them into a unified recommendation framework. The contextual factors such as individual preference and interpersonal influence work on the each users behaviors to adopt the recommended information. These two factors will highly provide an useruser interactions with their interpersonal influence or social relation which tells whether the user has close relation with item senders and individual preference which adopt the behavior of the user content whether the user likes the item or not. A novel probabilistic matrix factorization method is used in this social recommendation to fuse the user-user and user-item comparison and relations in a latent space. In order to analyze or operate the recommended contents we use effective scalable algorithm to find a probability of each user item content which shows a high valuable item content and it incrementally process the data. RELATED WORKSWe review several major approaches to recommendation methods. Collaborative filtering and content based filtering have been widely used to help users find out the most valuable information. With the emerge of social networks, researchers design trust-based (Ester) [7] and influencebased (J. Huang, 2010) [6] methods to take use of the power coming from user relationships for recommendation. Storing context data using data cubes, called context cubes, is proposed in (L. D. Harvel, 2004) [14] for developing context-aware applications that use archive sensor data. In this work, data cubes are used to store historical context data and to extract interesting knowledge from large collections of context data.(A. Karatzoglou, 2010) [4] Proposed a model based CF approach for making recommendation with respect to rich contextual information, namely multiverse recommendation. Specifically, they modeled the rich contextual information with item by Ndimensional tensor, and proposed a novel algorithm to make tensor factorization. (Soo, 2004) [11] Proposed a prototype design for building a personalized recommender system to recommend travel related information according to users' contextual information. (M.-H. Park, 2007) [9] Proposed a location based personalized recommender system, which can reflect users' personal preferences by modeling user contextual information through Bayesian Networks.(Hao Ma) [17] Analyse latent factor using probabilistic matrix factorization, we learn the user latent feature space and item latent feature space by employing a user soc...

show abstract

Recommender systems with social regularization

Cited by 1,380 publications

References 32 publications

User Adoption Tendency Modeling for Social Contextual Recommendation

User Adoption Tendency Modeling for Social Contextual Recommendation

A proximity measure for link prediction in social user-item networks

Scalable Recommendation With Social Contextual Information

Contact Info

Product

Resources

About