人工蜂群算法的改进及在空间数据聚类中的应用

人工蜂群算法是一种新兴的群智能算法,具有灵活、易于与其他技术结合、设置参数少、实用性强的优点,但同时也存在一些缺陷,如蜜源更新和选择机制随意性较大,算法后期收敛速度慢等。本文在总结前人研究成果的基础上,从蜜源更新公式和选择机制两方面对基本的人工蜂群算法做出改进。算法的应用是蜂群算法研究的热点,本文采用位置-属性一体化概念模型,将空间属性与非空间属性纳入统一的空间计算模型,结合本文提出的改进的人工蜂群算法对Meuse数据集进行空间聚类分析,获得了比较好的结果。

Improvement of Artificial Bee Colony Algorithm and Its Application in Spatial Data Clustering

Artificial bee colony algorithm is a new swarm intelligence algorithm,which is flexible,easy to combine with other technology,few parameters and practical advantages,but also has some defects,such as nectar update and selection of arbitrarily large,slow convergence speed and so on shortcomings.Based on the previous research results,this paper improves the basic artificial bee colony algorithm from the aspects of nectar regeneration formula and selection mechanism.The application of algorithms is a hot topic in the colony algorithm,this paper adopts the position attribute integration concept model,the spatial attributes and non-spatial attributes into a unified space calculation model,combined with the improved artificial bee colony algorithm is proposed for spatial clustering analysis on the Meuse data set,obtained good results.