Improved K-Means cluster algorithm in telecommunications enterprises customer segmentation

Abstract

According to the actual requirements of telecommunications enterprises customer segmentation, in this paper an improved K-Means algorithm was introduced. The algorithm was used to initialize cluster center in the cluster analysis phase of Data Mining technology. The experimental results proved that the novel algorithm has greater improvement than the original one in efficiency and accuracy. The results applying the novel method to telecommunications enterprises customer segmentation showed that the segmentation results obtained can be used as the data basis in differentiated services for customers and have positive significance for product design and phone packages recommendation.