一种新的混合蚁群聚类算法

A New Hybrid Ant Colony Algorithm for Clustering

针对蚁群聚类算法存在收敛速度慢、易陷入局部最优等缺陷,通过在蚁群聚类算法的每次迭代过程中引入遗传算法,提出一种混合蚁群聚类算法.它利用遗传算法全局快速收敛的特性,提升了蚁群聚类算法的收敛速度,同时,遗传算法中的交叉、变异操作扩大了解空间的搜索,帮助蚁群算法跳出局部最优.仿真试验验证了算法的性能.

Focusing on the problem that the ant colony clustering algorithm may be convergence slowly and easily fall into local optimal drawbacks. Proposed a new hybrid algorithm by adds GA to Ant Colony clus- tering algorithm＇s every generation. Making use of GA＇s advantage of whole quick convergence, Ant Colony clustering algorithm＇s convergence speed was improved. Meanwhile, the operation of crossover and mutation improved the ability of Ant Colony clustering algorithm to avoid being premature. This algorithm has been implemented and tested on several simulated datasets and UCI machine learning datasets. The authors＇ experiments reveal very encouraging results in terms of the quality of solution found and the processing time required.