山东理工大学学报(自然科学版)
山東理工大學學報(自然科學版)
산동리공대학학보(자연과학판)
Journal of Shandong University of Technology (Natural Science Edition)
2015年
5期
16-20
,共5页
K-means算法%粒子群算法%飞行时间因子%PSO-K算法
K-means算法%粒子群算法%飛行時間因子%PSO-K算法
K-means산법%입자군산법%비행시간인자%PSO-K산법
K-means algorithm%PSO algorithm%flight time factor%PSO-K algorithm
针对K‐means算法对初始聚类中心敏感、算法容易收敛于局部解等问题,运用了增加飞行时间因子的粒子群算法,提高粒子群算法性能。利用改进的粒子群算法与K‐means算法相结合,提高了基于粒子群算法的K‐means算法性能。数值试验验证了提出算法的收敛性,且最优解的精度优于K‐means算法、PSO算法和PSO‐K算法。
針對K‐means算法對初始聚類中心敏感、算法容易收斂于跼部解等問題,運用瞭增加飛行時間因子的粒子群算法,提高粒子群算法性能。利用改進的粒子群算法與K‐means算法相結閤,提高瞭基于粒子群算法的K‐means算法性能。數值試驗驗證瞭提齣算法的收斂性,且最優解的精度優于K‐means算法、PSO算法和PSO‐K算法。
침대K‐means산법대초시취류중심민감、산법용역수렴우국부해등문제,운용료증가비행시간인자적입자군산법,제고입자군산법성능。이용개진적입자군산법여K‐means산법상결합,제고료기우입자군산법적K‐means산법성능。수치시험험증료제출산법적수렴성,차최우해적정도우우K‐means산법、PSO산법화PSO‐K산법。
Considering K‐means algorithm was sensitive to the initial cluster centers and easy to converge to local solution and other issues ,we increased flight time factor to improve particle swarm algorithm performance .The improved particle swarm algorithm and K‐means algorithm were combined to improve the performance of K‐means algorithm based on particle swarm optimi‐zation .Numerical experiments verified the proposed convergence of the algorithm ,and the opti‐mal solution accuracy was better than K‐means algorithm ,PSO algorithm and PSO‐K algorithm .