K-means clustering algorithm based on Optimization of initial clustering center

Abstract:
K-means clustering algorithm based on initial cluster center optimization. Aiming at the fact that the running result of traditional K-means clustering algorithm depends on the initial cluster number and cluster center, this paper proposes a k-means algorithm based on initial cluster center optimization. According to the distribution characteristics of the data set, the algorithm selects the data far away as the initial clustering center, and determines the clustering number k value through the clustering compactness and quantifying the distance between samples. Experiments show that the improved clustering algorithm proposed in this paper obtains good clustering effect and high clustering accuracy.