An improved hierarchical clustering algorithm based on subtractive clustering

摘要

When dealing with large data sets, the hierarchical clustering algorithm generates a huge amount of computation and high spatial complexity. In order to quickly achieve better clustering under the premise of increasing the amount of data, this paper presents an improved algorithm of hierarchical clustering. Firstly, it uses subtractive clustering to get an initial clustering result, then, it uses the Kruskal minimum spanning tree-based algorithm to improve the hierarchical clustering algorithm to achieve clustering. Through simulation of the Colon dataset, the result showed improved hierarchical clustering based on the subtractive clustering algorithm which is faster and more efficient, and the cluster effect is also better than the traditional clustering algorithm.