一种优化初始中心的K—means聚类算法

针对传统K—means聚类算法对初始聚类中心的敏感性和随机性，造成容易陷入局部最优解和聚类结果波动性大的问题，结合密度法和最大化最小距离的思想，提出基于最近高密度点间的垂直中心点优化初始聚类中心的K—means聚类算法。该算法选取相互间距离最大的K对高密度点，并以这足对高密度点的均值作为聚类的初始中心，再进行K—means聚类。实验结果表明，该算法有效排除样本中含有的孤立点，并且聚类过程收敛速度快，聚类结果有更好的准确性和稳定性。

A K-means Clustering Algorithm of Meliorated Initial Center

The traditional K-means clustering algorithm has the sensitivity and randomness for initial clustering center. So it easily falls into local optimal solution and has unstable results. To solve the problem, proposed a K-means algorithm of meliorated initial clustering center based on vertical center point of the closest high density points. This algorithm selects K pairs of high density points that have the maximal distance between each other,and then uses the average values of K pairs of high density points as the initial clustering centers to implement the traditional K-means. The experimental results show that this algorithm is effective to eliminate isolated points and has better accuracy and stability.