An improved k-means clustering collaborative filtering recommendation algorithm
Xiaoying Ye, Rong Tang · 2022 5th International Conference on Advanced Electronic Materials, Computers and Software Engineering (AEMCSE) · 2022
Aiming at the problem of low recommendation quality and low recommendation efficiency caused by sparse data of collaborative filtering algorithm, a collaborative filtering recommendation algorithm based on improved k-means clustering is proposed. The algorithm first uses canopy algorithm to roughly cluster the data, uses the maximum and minimum distance product method to select the initial point, and improves the randomness of the initial center selection of K-means algorithm. After generating multiple clusters, the modified cosine similarity is combined with user attribute features to form a new similarity calculation model and realize the corresponding recommendation. Simulation results show that the Mae and RMSE values of this algorithm are significantly better than other comparison algorithms. The improved algorithm can improve the recommendation efficiency and accuracy.