Abstract | ||
---|---|---|
Collaborative Filtering (CF) is a popular way to build recommender systems and has been widely deployed by many e-commerce websites. Generally, there are two parallel research directions on CF, one is to improve the prediction accuracy ~ (i.e., effectiveness) of CF algorithms and others focus on reducing time cost of CF algorithms ~ (i.e., efficiency). Nevertheless, the problem of how to combine the complementary advantages of these two directions, and design a CF algorithm that is both effective and efficient remains pretty much open. To this end, in this paper, we provide a Matrix Factorization based on Co-Clustering (MFCC) algorithm to address the problem. Specifically, we first adopt a co-clustering algorithm to cluster the user-item rating matrix into several separate sub rating matrices. After that, we provide an efficient matrix factorization algorithm by utilizing the strong connections of users and items in each cluster. In the meantime, this process is also efficient as we can simultaneously compute the matrix factorization for each cluster as there exists little interactions among different clusters. Finally, the experimental results show both the effectiveness and efficiency of our proposed model. |
Year | Venue | Field |
---|---|---|
2017 | ICIMCS | Recommender system,Mel-frequency cepstrum,Cluster (physics),Collaborative filtering,Existential quantification,Pattern recognition,Matrix (mathematics),Computer science,Matrix decomposition,Theoretical computer science,Artificial intelligence,Biclustering |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
14 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wenjuan Yang | 1 | 0 | 2.70 |
Le Wu | 2 | 252 | 33.83 |
Xueliang Liu | 3 | 76 | 15.56 |
Chunxiao Fan | 4 | 0 | 0.68 |