Optimization Of Bee Colony Algorithm And Its Application In Clustering

With the rapid development of information technology,mobile Internet is widely popularized,and a large amount of rapidly updated data increases the complexity of practical problems.How to perform big data analysis efficiently and quickly has attracted widespread attention.Big data research aims to search for valuable information from a large amount of dynamically changing data,such as data analysis,data classification,prediction,and descriptive statistics.The swarm intelligence algorithm is one of the algorithms that can effectively assist in solving big data problems.When dealing with large and complex problems,the swarm intelligence algorithm,as an intelligent algorithm inspired by collaborative cooperation among populations,can propose effective and high-quality solutions.Cluster analysis technology gathers data with similar characteristics in the same cluster class,and stores data with large differences in characteristics in different categories,which belongs to an efficient data aggregation technology.Combining swarm intelligence algorithms with clustering algorithms can solve high-dimensional data analysis problems more effectively,and at the same time researched improved algorithms to improve the performance of original algorithms and solve data processing problems in the real world.This paper uses the ability of the artificial bee colony algorithm to quickly search globally.In this paper,the optimized bee colony algorithm is applied to improve the density peak clustering algorithm,and the artificial bee colony algorithm is used to search for the optimal solution of the target.The main contents and innovations of this article are as follows:(1)An efficient universal bee colony optimization algorithm(EUBCOA)was proposed.The algorithm proposes two improved methods.First,the search factor u is introduced,and the following bee selection strategy based on the local optimal solution is adopted.In order to realize the controllability of algorithm search ability,the search factor u is introduced to improve the global search range and local search range.In the later stage of the iteration,the following bee selection strategy based on the local optimal solution is adopted to improve the optimization rate of the algorithm to find the target optimal solution.The algorithm selects 10 Benchmark functions for simulation optimization test.The experimental results show that the improved algorithm improves the rate of finding the optimal value of the target,and confirms the efficiency of the EUBCOA algorithm in processing optimization problems.(2)Bee colony algorithm based on adaptive intensity adjustment factor(CAIBC)was proposed.In order to improve the search accuracy and search rate,a new adaptive update step size Li is introduced.The adaptive update step size is introduced to increase the step size in the early search period and expand the search range.With the iterative update of the nector,the local search is refined to improve the accuracy of the algorithm.In this paper,six Benchmark functions are selected to test the optimization performance of the CAIBC algorithm.The experimental results and data confirm that the CAIBC algorithm can effectively improve the target optimal solution quality of the algorithm.(3)Improved density peak clustering algorithm based on artificial bee colony(IDPCA)was proposed.Firstly,the random direction vector factor is introduced to comprehensively collect data information.Secondly,the improved artificial bee colony algorithm is used to obtain the cluster center and optimize the category of the data points.In order to confirm the efficient data analysis ability of the IDPCA algorithm proposed in this paper,this paper selects 9 data sets and uses the Silhouette and F-measure indicators to test the clustering performance of the IDPCA algorithm.On this basis,IDPCA algorithm was applied to the economic field,and good experimental results were obtained.