摘要
为提高K-means聚类算法在异常检测中的效果,以聚类分析为主线,针对传统的K_means聚类算法在初始聚类中心点选择随机性和K值预先设定的问题,提出了一种改进的K_means聚类分析算法,算法引入密度参数和距离理论。依据密度理论和最大距离找出k个初始化中心点。并对算法进行仿真实验,实验证明,新的算法具有良好的效果。
In order to improve K-means clustering algorithm effect in Anomaly Detection,use Cluster analysis to the main line,for K means clustering algorithms the problems in the initial cluster centers select random and the preset value K. an algorithm to calculate the number of the Cluster Center was given.Proposed an improved K-means Clustering Algorithm,we used Density parameter and the theoretical of distance in this algorithm. according to the theoretical density and maximum distances to find k initialization center. For This Algorithm ,we set to test, The results show the algorithm have a higher detection rate and a lower false alarm rate.
出处
《电子技术（上海）》
2016年第11期30-32,共3页
Electronic Technology
