基于混合粒子群和蚁群算法融合的聚类算法  

Clustering Algorithm Based on Combination of Hybrid Particle Swarm and Ant Colony Algorithm

作　　者：胡人远 张之明[2] HU Renyuan ZHANG Zhiming(Postgraduate Brigade Department of Information Engineering, Engineering University of PAP, Xi＇an 710086, China)

机构地区：[1]武警工程大学研究生管理大队,西安710086 [2]武警工程大学信息工程系,西安710086

出　　处：《武警工程大学学报》2016年第6期15-19,共5页Journal of Engineering University of the Chinese People＇s Armed Police Force

基　　金：国家自然科学基金项目“多层多域智能光网络安全路由算法与协议研究”（61402529）

摘　　要：针对基于遗传与蚁群算法融合的聚类算法（Clustering Algorithm based on Genetic and Ant Colony Algorithm，GA2C2A）存在局部搜索能力差、易陷入局部最优及遗传算法操作复杂的问题，提出了基于混合粒子群和蚁群算法融合的聚类算法（Clustering Algorithm based on Hybrid Particle Swarmand Ant Colony Algorithm Integration，HPS0-ACA）。算法首先针对标准粒子群算法存在的缺陷进行了改进，形成了局部搜索能力强、操作实现简单的混合粒子群算法，并用来代替遗传算法与蚁群算法结合。以加州大学的3种数据集为对象进行数据聚类实验。结果表明，与基于遗传与蚁群算法融合的聚类算法相比，该算法在聚类性能上有明显的优势。To resolve the shortage that the GA2C2A exists problems such as poor local search ability,easines to fall into local optimum and complex genetic algorithm operation, we pro- pose HPSO-ACA. Aiming at the existing defects of PSO, the algorithm proposed formed a hybrid particle swarm optimization algorithm with strong local search ability and simple op- eration in place of the integration of genetic algorithm and PSO. And In the data clustering experiments on three data sets of UCI data sets by the algorithm, the experimental results show that, compared with GA2C2A, the algorithm has obvious advantage on the clustering performance.

关 键 词：数据挖掘 粒子群算法 混合粒子群算法 云模型 蚁群算法 聚类 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]