Improvement And Application Of K-means Clustering Algorithm Based On Hybrid Frog Leaping Algorithm

Now is a rapid development of science and technology,information flow fast society.The communication of people become more and more close and life become more and more convenient,each field is filled with various types of data.To extract useful information from these data,we need to analyze the data.In the data mining technology,clustering analysis is one of its analytical tools,and has been widely concerned in the field of data mining,has been widely used in various fields,including image recognition,web search,information security,business intelligence and so on.K-means clustering algorithm is a kind of classical clustering algorithm based on partitioning technology.It has the characteristics of principle easy to understand,fast convergence speed and easy implementation.And it is more efficient when using K-means clustering for large-scale data.So it has been more widely used in the field of data mining.Due to the continuous innovation and development of data mining technology,many scholars have also applied the intelligent optimization algorithm to K-means.Hybrid leapfrog algorithm is a new intelligent optimization method in the field of evolutionary computation.It is widely used in many fields because of its easy understanding of theory,less adjustment parameters,high calculation rate,strong global optimization ability and easy implementation.It has achieved gratifying research results and has become a hot and one of the key in the field of artificial intelligence.Because K-means clustering algorithm is more sensitive to the selection of initial clustering centroids and different initial clustering centroids tend to cause different clustering results,this paper presents a K-means combining algorithm based on hybrid frog leaping algorithm.The specific work of this paper is as follows:1.The concepts of data mining,clustering analysis,K-means and hybrid leapfrog algorithm are briefly introduced.The algorithmic process,similarity measure method and advantages and disadvantages of the two algorithms are analyzed.2.An improved hybrid leapfrog algorithm is proposed.The idea of random grouping is introduced for the hybrid leapfrog algorithm.With the evolution of each iteration of the population,a random grouping is carried out to achieve the goal ofgroup optimization firstly and then population optimization.And the algorithm can adjust the moving step size of local optimization by introducing the inertia coefficient,which further prevents the possibility of the algorithm falling into the local optimal solution and improves the global optimization ability of the hybrid leapfrog algorithm.3.The improved hybrid leapfrog algorithm is applied to the improved K-means,and the performance of the original algorithm is improved by using the advantages of the two improved algorithms in the data set.In this paper,we optimize the optimal solution by choosing each data in the data set to be a frog and the intra-cluster variation is taken as the objective function.Finally,the clustering analysis of K-means clustering algorithm under the condition of initial cluster centroid is finished.4.An improved K-Means of hybrid frog leap algorithm based on random grouping is used to calculate the data of the five indicators of GDP,child care ratio,old age dependency ratio,population birth rate and population mortality in the last two decades of China to carry out data clustering analysis.According to the clustering results,the correlation analysis of each category is carried out,and the correlation between the five indexes is analyzed,which indicates that the algorithm has a good application prospect.