基于阈值和蚁群算法结合的聚类方法

Clustering Method Combining Threshold Algorithm with Ant Colony Algorithm

为了改善聚类分析的质量,提出了一种基于阈值和蚁群算法相结合的聚类方法.按此方法,首先由基于阈值的聚类算法进行聚类,生成聚类中心,聚类个数也随之初步确定;然后将蚁群算法的转移概率引入K-平均算法,对上述聚类结果进行二次优化.实验表明,与K-平均算法等相比,该聚类方法的F-测度值(F-m easure)更高.

To improve the quality of clusting analysis, a novel clustering method combining the threshold algorithm with the ant colony algorithm was proposed. With this method, the center and number of clustering are determined by using the clustering algorithm based on threshold, and then the above clustering results are optimized by the K-means algorithm combining with transition probability based on the ant colony algorithm. The experimental results show that the proposed clustering method has a higher F-measure than the K-means and other algorithms.