Social Recommendation With Evolutionary Opinion Dynamics

Xiong, Fei; Wang, Ximeng; Pan, Shirui; Yang, Hong; Wang, Haishuai; Zhang, Chengqi

doi:10.1109/tsmc.2018.2854000

Cited by 67 publications

(46 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…. Social network-based recommender systems [22] can mitigate the cold-start problem, and most of the information on social networks is unreliable due to different backgrounds or preferences of users. e trust network is introduced into the recommender system to solve the problem of a fraud attack.…”

Section: Trust Recommender Systemmentioning

confidence: 99%

A Hierarchical Attention Recommender System Based on Cross-Domain Social Networks

Zhao

Xiong

et al. 2020

Complexity

View full text Add to dashboard Cite

Search engines and recommendation systems are an essential means of solving information overload, and recommendation algorithms are the core of recommendation systems. Recently, the recommendation algorithm of graph neural network based on social network has greatly improved the quality of the recommendation system. However, these methods paid far too little attention to the heterogeneity of social networks. Indeed, ignoring the heterogeneity of connections between users and interactions between users and items may seriously affect user representation. In this paper, we propose a hierarchical attention recommendation system (HA-RS) based on mask social network, combining social network information and user behavior information, which improves not only the accuracy of recommendation but also the flexibility of the network. First, learning the node representation in the item domain through the proposed Context-NE model and then the feature information of neighbor nodes in social domain is aggregated through the hierarchical attention network. It can fuse the information in the heterogeneous network (social domain and item domain) through the above two steps. We propose the mask mechanism to solve the cold-start issues for users and items by randomly masking some nodes in the item domain and in the social domain during the training process. Comprehensive experiments on four real-world datasets show the effectiveness of the proposed method.

show abstract

Section: Trust Recommender Systemmentioning

confidence: 99%

A Hierarchical Attention Recommender System Based on Cross-Domain Social Networks

Zhao

Xiong

et al. 2020

Complexity

View full text Add to dashboard Cite

show abstract

“…e combination of network distance and clustering coefficient can demonstrate the "small world" effect, showing how nodes are embedded in their surrounding nodes. By referring to relevant research results [37][38][39], an indicator is proposed here to combine the two, as shown in (2), and the last column of Cd is obtained as in Table 2; the higher the Cd, the more obvious the "small world" effect.…”

Section: Complexitymentioning

confidence: 99%

Analysis of the Structural Characteristics of the Online Social Network of Chinese Professional Athletes

et al. 2021

View full text Add to dashboard Cite

In response to the lack of research on the online social network structure of athletes, elements of research on the online social network structure of athletes were constructed based on the whole network perspective and through the study of the characteristics of the whole online social network structure of athletes, in order to provide reference for the physical and mental health development of athletes from a new perspective. Data were collected through questionnaires, and several software programs were used to preprocess and analyse the collected data. Through the analysis of the online whole network structure, it was found that the network density of the online support network was generally greater than that of the online discussion network, and athletes still showed stronger practical support demands and behaved more rationally in the process of training and learning life, while from the perspective of the relationship structure, the athletes’ family and classmates’ online support is weaker than that in previous studies; in terms of the whole network, strong relationships still dominate in this population, while attention should be paid to the impact of weak relationships.

show abstract

“…we use the gradient descent to reach a local minimization of the loss function. Gradient descent is an effective way for minimization when objective functions are differentiable and non-convex, and is also the most commonly used algorithm in MF-based recommender systems [18,24,26,38,40,45]. The gradients of the parameters Θ are performed as follows:…”

Section: Learning and Predictionmentioning

confidence: 99%