关键词: 推荐系统, 决策, 协同过滤, 中心聚集参数, K-means聚类

Abstract: Collaborative filtering recommendation algorithm is widely used in personalized recommendation system. However, it has some shortcomings in data sparseness and scalability. This paper proposes a novel approach to solve the problems, which contains two components. The first component is to solve the problem of data sparseness by filling the initial scoring matrix with missing values based on Slope One algorithm and using the collaborative filtering algorithm based on K-means clustering to predict the score of target users. The second one is to solve the scalability problem by proposing an improvedimproveu K-means algorithm based on central aggregation parameters. The results of the relevant comparative experiments on Movielens dataset show that the recommended precision is significantly improved and the data sparseness and scalability issues are effectively improved. Therefore, the research conclusions of this paper can not only further enrich the existing theoretical achievements of collaborative filtering recommendation algorithm, but also provide theoretical basis and decision-making reference for improving the accuracy of the recommendation system.

Key words: recommendation system, decision making, collaborative filtering, central aggregation parameter, K-means clustering