TY - GEN
T1 - SOM clustering collaborative filtering algorithm based on singular value decomposition
AU - Ma, Xiaopan
AU - Li, Xiaojing
AU - Guo, Dong
AU - Cui, Lixin
AU - Jiang, Xuru
AU - Chen, Xin
N1 - Publisher Copyright:
© 2019 Association for Computing Machinery.
PY - 2019/4/12
Y1 - 2019/4/12
N2 - The application of traditional collaborative filtering algorithm on large-scale commercial websites is very mature. However, the data sparsity and extensibility problems that occur in the algorithm affect the recommendation accuracy of the algorithm. In order to solve this problem, a SOM clustering collaborative filtering algorithm based on singular value decomposition is proposed. Firstly, the original sparse matrix is reduced by the singular value decomposition, and the items are evaluated in the low-dimensional space, the prediction results are filled in the original matrix, which alleviates the problem of data sparseness. Then use SOM to cluster the users, which reduces the range of users searching for neighbors and improves the scalability of the algorithm. The experimental results on MovieLens-100k show that the algorithm can effectively improve the accuracy of the recommendation.
AB - The application of traditional collaborative filtering algorithm on large-scale commercial websites is very mature. However, the data sparsity and extensibility problems that occur in the algorithm affect the recommendation accuracy of the algorithm. In order to solve this problem, a SOM clustering collaborative filtering algorithm based on singular value decomposition is proposed. Firstly, the original sparse matrix is reduced by the singular value decomposition, and the items are evaluated in the low-dimensional space, the prediction results are filled in the original matrix, which alleviates the problem of data sparseness. Then use SOM to cluster the users, which reduces the range of users searching for neighbors and improves the scalability of the algorithm. The experimental results on MovieLens-100k show that the algorithm can effectively improve the accuracy of the recommendation.
KW - Collaborative filtering algorithm
KW - Recommendation algorithm
KW - Self organizing map
KW - Singular value decomposition
UR - https://www.scopus.com/pages/publications/85068856614
U2 - 10.1145/3325730.3325740
DO - 10.1145/3325730.3325740
M3 - Conference contribution
AN - SCOPUS:85068856614
T3 - ACM International Conference Proceeding Series
SP - 61
EP - 65
BT - ICMAI 2019 - Proceedings of 2019 4th International Conference on Mathematics and Artificial Intelligence
PB - Association for Computing Machinery
T2 - 4th International Conference on Mathematics and Artificial Intelligence, ICMAI 2019
Y2 - 12 April 2019 through 15 April 2019
ER -