摘要
针对粗糙K-均值聚类算法中类均值计算式的特点,提出了一种改进的粗糙K-均值算法。改进后的算法基于数据对象所在区域的密度,在类的均值计算过程中对每个对象赋以不同的权重。不同测试数据集的实验结果表明,改进后的粗糙K-均值算法提高了聚类的准确性,降低了迭代次数,并且可以有效地减小孤立点对聚类的影响。
According to the feature of the calculation of means in Rough K-means algorithm, an improved Rough K- means algorithm was proposed. The new algorithm introduces weights to the calculation of means, which is based on the density of each point. The experiments show that the new algorithm improves the clustering accuracy and reduces the iteration times as well as the outliers＇ influence.
出处
《计算机科学》
CSCD
北大核心
2009年第3期220-222,共3页
Computer Science
基金
国家自然科学基金项目(60775036
60475019)
高等学校博士学科点专项科研基金(20060247039)资助
关键词
聚类算法
粗糙K-均值
密度
孤立点
Clustering algorithm, Rough K-means, Density, Outlier
