A New Selection Method of K-means Clustering with Initial Clustering Center Point

Clustering analysis has a wide range of applications in many areas, such as information retrieval and data mining, in which k-means algorithm is a more succinct and faster clustering algorithm. But there are also some defects. The initial clustering number must be predetermined in k-means clustering algorithm, and the choice of the initial center has randomness too. To the shortcoming of appointing the initial center at random in k-means clustering algorithm, this paper proposes a new method of k-means algorithm with refined initial center point based on data sample distribution. Experiment reveals that the clustering results produced by this algorithm are more stable and have high quality.