Improvement Study and Application Based on K-Means Clustering Algorithm

The k-means algorithm is the most classic cluster algorithm based on the distance in the Cluster-analysis. Because of the shortcomings of the traditional k-means algorithm, this paper proposes an improved k-means al- gorithm which is researched and analyzed. The results, putting into this im- proved algorithm to analyze the data of a shopping mall, prove that the algorithm can improve the quality of cluster algorithm and attain a rather good result.