Staged text clustering algorithm based on K-means and hierarchical agglomeration clustering

Abstract

In order to achieve better clustering, a staged hybrid clustering algorithm SC-KH (Staged text clustering algorithm based on K-means and HAC) is proposed to overcome the shortcomings of low accuracy of K-means algorithm and high time complexity of hierarchical agglomeration clustering algorithm. The algorithm divides the clustering process into two stages: splitting and merging. The K-means algorithm is used in the splitting stage, and the k value and the initial clustering center are provided using the Canopy algorithm. The hierarchical agglomeration clustering algorithm is used in the merging stage. The experimental results show that the performance of the improved SC-KH algorithm is better than that of K-means and hierarchical agglomeration, and the clustering quality is improved.