K-means Initial Clustering Center Optimal Algorithm Based on Partitioning
Zhan Hai-liang · Microelectronics & Computer · 2009
A new method of finding the initial center of K-means algorithm is given.It first estimates the cluster center;and then initializes the K-means algorithm with the estimated center.The experiment demonstrates that the new method can produce high quality cluster result and is superior to the initialization method of cluster centers.