基于改进的SVD算法和二分K-均值聚类算法的协同过滤算法
Collaborative filtering algorithm based on the improved SVD algorithm and binary K-means clustering algorithm
-
摘要: 针对传统的协同过滤算法存在稀疏性较大和扩展性较差的问题,提出了基于改进的奇异值分解(SVD)算法和二分K-均值聚类算法的协同过滤算法.该算法首先利用改进的SVD算法对稀疏的用户-项目评分矩阵进行降维,获得用户隐含特征矩阵,然后运用二分K-均值聚类算法对相似用户进行聚类来提升算法的可扩展性,最后利用最近邻居集的评分修正目标用户的评分,以减小因矩阵分解导致用户信息丢失造成的误差.利用MovieLens 100K数据集进行的实验结果表明,与传统的基于用户的协同过滤算法、基于K-均值聚类的协同过滤算法和隐语义模型(LFM)算法相比,本文提出的算法能够有效提高推荐结果的准确性.
-
关键词:
- 个性化推荐 /
- SVD算法 /
- 二分K-均值聚类算法 /
- 协同过滤 /
- 矩阵分解
Abstract: Aiming at the problem of large sparseness and poor scalability of traditional collaborative filtering algorithms,a collaborative filtering algorithm based on the improved singular value decomposition (SVD) algorithm and binary K-means clustering algorithm was proposed.The algorithm firstly used the improved SVD algorithm to reduce the dimensionality of the sparse user-item rating matrix to obtain the user implicit feature matrix,then used the binary K-means clustering algorithm to cluster similar users to improve the scalability of the algorithm,and finally used the nearest neighbor set score to correct the target user's score to make up for the error caused by the loss of user information due to matrix factorization.Experimental results on the MovieLens 100K data set showed that compared with the traditional user-based collaborative filtering algorithm,K-means clustering-based collaborative filtering algorithm and latent factor model (LFM) algorithm,this method could effectively improve the accuracy of recommendation results. -
-
[1]
SUN P,LI Z,HAN Z,et al.An overview of collaborative filtering recommendation algorithm[J].Advanced Materials Research,2013,756-759:3899.
-
[2]
张琳,闫强.基于管理和消费者行为视角的个性化推荐研究与展望[J].北京邮电大学学报(社会科学版),2016,18(6):24.
-
[3]
BRUSILOVSKY P,KOBSA A,NEJDL W.The adaptive web:Methods and strategies of web personalization[J].Lecture Notes in Computer Science,2007,2002(5):377.
-
[4]
LI B,ZHU X Q,LI R J,et al.Cross-domain collaborative filtering over time[C]//Proceedings of the 22nd International Joint Conference on Artificial Intelligence.Palo Alto:AAAI Press,2011:2293.
-
[5]
SUGANESHWAR G,IBRAHIM S P.A survey on collaborative filtering based recommendation system[C]//3rd International Symposium on Big Data and Cloud Computing.Chengdu:Springer-Verlag,2016:503.
-
[6]
KOOHI H,KIANI K.User based collaborative filtering using fuzzy C-means[J].Measurement,2016,91(1):134.
-
[7]
SU X,KHOSHGOFTAAR T M.Collaborative filtering for multi-class data using belief nets algorithms[J].International Journal on Artificial Intelligence Tools,2008,17(1):71.
-
[8]
孙小华,陈洪,孔繁胜.在协同过滤中结合奇异值分解与最近邻方法[J].计算机应用研究,2006,23(9):206.
-
[9]
刘艺,冯钧,魏童童,等.一种改进的协同过滤推荐算法[J].计算机与现代化,2017(1):1.
-
[10]
SARWAR B,KARYPIS G,KONSTAN J,et al.Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the 10th International Conference on World Wide Web.New York:ACM,2001:285.
-
[11]
PANDEY A,PANDEY R.Elective recommendation support through K-means clustering using R-tool[C]//International Conference on Computational Intelligence and Communication Networks.Kolkata:ICRCICN,2016:851.
-
[12]
HARPER F M,KONSTAN J A.The MovieLens datasets:History and context[J].ACM Transactions on Interactive Intelligent Systems,2015,5(4):1.
-
[13]
陈清洁.基于SVD的协同过滤推荐算法研究[D].成都:西南交通大学,2017.
-
[1]
计量
- PDF下载量: 8
- 文章访问数: 945
- 引证文献数: 0