Abstract | ||
---|---|---|
Social recommendation is the recommendation process by using social information. Being created to annotate and categorize items, tags are shared by users to find, organize and understand online entities, and provide the inherent flexibility to predict users' preference. In this paper, tag information and social relations are combined into recommendation, and a new tag information aided social recommendation method is proposed. By means of matrix factorization and regularization technology, we calculate the feature vectors and predict the rating of users. Also a new similarity measuring method is put forward when exploiting tag information through link prediction. Experiments show that our method generates better performance than state-of-the-art methods. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/DSC.2016.30 | 2016 IEEE First International Conference on Data Science in Cyberspace (DSC) |
Keywords | Field | DocType |
Social Recommendation,Tag Information,Matrix Factorization,Regularization | Social relation,Categorization,Data mining,Feature vector,Information retrieval,Computer science,Matrix decomposition,Side information,Prediction algorithms,Linear programming,Social information | Conference |
ISBN | Citations | PageRank |
978-1-5090-1193-3 | 0 | 0.34 |
References | Authors | |
0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Xiang Hu | 1 | 0 | 0.34 |
Wendong Wang | 2 | 821 | 72.69 |
Xiangyang Gong | 3 | 161 | 23.01 |
Bai Wang | 4 | 2 | 1.18 |
Xirong Que | 5 | 142 | 15.76 |
Hongke Xia | 6 | 0 | 1.35 |