| 融合K-调和均值和模拟退火粒子群的混合聚类算法 |
| |
| 引用本文： | 毛力,刘兴阳,沈明明.融合K-调和均值和模拟退火粒子群的混合聚类算法[J].计算机与应用化学,2011,28(2). |
| |
| 作者姓名： | 毛力 刘兴阳 沈明明 |
| |
| 作者单位： | 中国水产科学研究院淡水渔业研究中心,农业部淡水鱼类遗传育种和养殖生物学重点开放实验室江南大学信息工程学院,江苏,无锡,214122 |
| |
| 基金项目： | 农业部淡水鱼类遗传育种和养殖生物学重点开放实验室开放基金资助项目(BZ2009-07) |
| |
| 摘 要： | 针对K-调和均值和模拟退火粒子群聚类算法的优缺点,提出了1种融合K-调和均值和模拟退火粒子群的混合聚类算法。首先通过K-调和均值方法将粒子群分成若干个子群,每个粒子根据其个体极值和所在子种群的全局极值来更新位置。同时引入模拟退火思想,抑制了早期收敛,提高了计算精度。本文使用Iris、Zoo、Wine和Image Segmentation,4个数据库,以F-measure为评价聚类效果的标准,对混合聚类算法进行了验证。研究发现,该混合聚类算法可以有效地避免陷入局部最优,在保证收敛速度的同时增强了算法的全局搜索能力,明显改善了聚类效果。该算法目前已用于无锡一淡水养殖基地的水产健康养殖水质分析系统,运行效果良好。
|
| 关 键 词： | K-调和均值 模拟退火 粒子群算法 聚类 |
A hybrid clustering algorithm combining K-harmonic means and simulated annealing particle swarm optimization |
| |
| Mao Li,Liu Xingyang,Shen Mingming.A hybrid clustering algorithm combining K-harmonic means and simulated annealing particle swarm optimization[J].Computers and Applied Chemistry,2011,28(2). |
| |
| Authors: | Mao Li Liu Xingyang Shen Mingming |
| |
| Affiliation: | Mao Li~(1,2),Liu Xingyang~(1,2) and Shen Mingming~(1,2) (1.Key Laboratory of Genetic Breeding and Aquaculture Biology of Freshwater Fishes,Ministry of Agriculture,Freshwater Fisheries Research Center,Chinese Academy of Fishery Science,Wuxi,214081,Jiangsu,China) (2.School of Information Technology,Jiangnan University,214122,China) |
| |
| Abstract: | In view of the advantages and disadvantages of K-harmonic means(KHM) and simulated annealing particle swarm optimization (SAPSO),a hybrid clustering algorithm combining KHM and SAPSO(KHM-SAPSO) was presented in this paper.With KHM,the particle swarm was divided into several sub-groups.Each particle iteratively updated its location based on its individual extreme value and the global extreme value of the sub-group it belonged to.With simulated annealing technique,the algorithm prevented premature convergence... |
| |
| Keywords: | K-harmonic means simulated annealing particle swarm optimization cluster |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
