An asymmetric graph recommendation algorithm based on user scores and interests

2017 2nd International Seminar on Advances in Materials Science and Engineering

Zhang, Yi-Wen^1 ; Wang, Ran^2 ; Wang, Ting-Wei^1 ; He, Guang-Da^1 ; Hu, Jun^1

An Hui Xin Hua University, Hefei, China

230088, China^1

An Hui University, Hefei, China

230601, China^2

关键词: Common projects;    Prediction accuracy;    Recommendation accuracy;    Recommendation algorithms;    Similarity calculation;    User rating;    User's interest;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/231/1/012045/pdf
DOI  :  10.1088/1757-899X/231/1/012045

来源: IOP

【 摘 要 】

The proposed algorithm uses the symmetry similarity calculation method based on the common project, which can not reflect the user's interest very well and the recommendation accuracy is low. In order to improve the accuracy of recommendation, this paper proposes a graph recommendation algorithm which integrates user rating and interest. In order to reflect the interests of users, using the principle of diffusion of the substance user rating information asymmetry; Experiments show that the proposed algorithm for user rating and interest fusion improves the prediction accuracy.