Application of improved k-means k-nearest neighbor algorithm in the movie recommendation system

Abstract

In this paper, we propose a clustering and reclassification method for movie recommendation. We use the improved K-means algorithm to cluster according to the scores of similar users, Firstly, the elbow function is used to estimate the number of clusters, and the elbow method is used to determine the K value. Then, the K-means algorithm of the maximum and minimum distance method is used to select the initial cluster center, and finally the cluster and cluster center are obtained. According to the similarity between the test data of the user's rating and user's personal information and the clustering center, they are divided into the cluster to which they belong, and the sample set in the cluster is used as the training set for K-nearest neighbor classification.