Slope One Recommendation Algorithm Based on User Clustering and Scoring Preferences
Yue Song, Wu Sheng · Procedia Computer Science · 2020
The Slope One algorithm in the recommendation system has the characteristics of real time efficiency, convenient operation, but at the same time there are not considering between project and the similarity between user’s question. In order to improve such problems and improve the accuracy of the algorithm, this paper proposes a Slope One recommendation algorithm that integrates user clustering and scoring preference. First, the similarity among users is measured, and the improved K-means++ algorithm is used to divide users into several categories according to the similarity degree of project preferences. Then, in the category of target users, the Slope One algorithm that integrates users’ rating preferences is used to predict the score of projects. And finally top-n recommendation is made according to the predicted score. In this paper, using the Movielens dataset to experiment, the results show that the proposed algorithm can effectively reduce the mean absolute error and root mean square error of traditional algorithm, recommend have higher accuracy and better recommendation quality.