Improved K-Means Algorithm and Application in Customer Segmentation

Nowadays, clustering algorithms are widely used in the commercial field, such as customer analysis, and this application has achieved good effect. K-means algorithm is by far the most commonly used method for clustering. Although, the time consumption is fairly high when faced with lager-scale data. In this paper, we improved the K-means algorithm. Our improvement is based on the triangle inequality theorem. We use the improved algorithm to carry out a case study in the customer classification. The experimental results show that the improved method indeed lead to lower time consumption, and therefore more effective for large-scale dataset.