一种MDHD-K-means算法的研究

Study of a MDHD-K-means Algorithm

本文介绍了广泛应用的k-means算法。针对其选择初始聚类中心和易受孤立点的影响的缺点,给出了改进的算法。首先使用距离法移除孤立点,然后采用最大高密度距离法(MDHD)对初始聚类中心的选择进行改进。并进行了改进前后的对比实验。实验结果表明,改进后的算法比较稳定、准确。改进后的算法应用到计算机语言教学中,达到较好的分类效果。

The classic algorithm of k-means is discussed, that is one of the most widespread methods in clustering including both strongpoints and shortages. Not only is it sensitive to the original clustering center, but also it may be affected by the outliers. Given these shortages, an improved algorithm is discussed, which makes improvements in outliers and selection of original clustering center. The outlier detection based on the distance method. To select original clustering center based on the max-distance and high-density. This paper presents the application which all show that the improved algorithm can lead to better and more stable solutions than k-means algorithm. The experiment and application affection by the outliers is down to a much low fi gure. The improved algorithm was used to the teaching of computer applied in student achievement.