Improvement Of Artificial Bee Colony Algorithm And Its Application In Spatial Clustering

Artificial bee colony(ABC)algorithm is a new kind of swarm intelligence algorithm,with flexible,easy to combine with other technologies,set a few parameters,practical advantages and is widely used in management science,control engineering,image processing,data mining and other fields.ABC algorithm is a younger algorithm in swarm intelligence algorithm,which has many advantages and some flaws,such as the randomness of nectar update and selection mechanism,and the late convergence speed of the algorithm.On the basis of summarizing the previous research results,this paper improves the basic ABC algorithm from the aspects of the regeneration formula and the selection mechanism of the onlooker's stage,then the ABC algorithm is extended to the Artificial bee colony clustering(CABC)algorithm,afterwards a hybrid clustering algorithm based on the improved artificial bee colony clustering algorithm(ICABC)and the FCM algorithm is proposed.The application of algorithms is a hot topic to the ABC algorithm,in this paper,the CABC algorithm,the ICABC algorithm and the hybrid algorithm are used into the spatial clustering analysis of the Meuse data set and some county economic data sets in China and carried out with good results.The main work of this paper is as follows:Firstly,this paper analyzes the research background and significance of this paper,analyzes the research progress at home and abroad from the aspects of spatial clustering analysis and the ABC algorithm,and points out the main problems in the current research.Secondly,this paper introduces the related theoretical and technical basis,including the related contents of clustering analysis and spatial clustering analysis,fuzzy set theory and fuzzy clustering,and the computational mechanism and characteristics of swarm intelligence algorithm.Thirdly,summarizing the methods to improve the ABC algorithm,and then improves the ABC algorithm from the aspects of nectar regeneration formula and selection mechanism,after introducing five commonly used test functions to compare the performance of all the algorithms by experiment.Frothily,extending the ABC algorithm to the CABC algorithm,aiming at the characteristics of the FCM algorithm and the ICABC algorithm,a hybrid clustering algorithm based on the FCM algorithm and the ICABC algorithm is proposed,and the process of algorithm is expatiated in detail.Finally,verifying the effectiveness of the algorithm through five UCI data sets.Fifth,putting the CABC algorithm,the ICABC algorithm and the hybrid clustering algorithm into spatial clustering,the experimental results show that the results of the algorithm clustering are in good agreement with the actual situation,in terms of the convergence rate,accuracy,optimization precision and stability,the ICABC algorithm and the hybrid clustering algorithm are better than the CABC algorithm.