With the increasing popularity of computer technology and networks,it is inevitable that a large amount of data will be generated.When the amount of generated data is large,it is a matter of concern to obtain useful data information in a huge data ocean.Relying on traditional database queries is becoming more and more difficult,and the data that is queried is often not satisfactory.Furthermore,there are more and more technologies that require data calculations such as big data,data mining,and cloud computing.Effective data information can help managers and decision makers make correct decisions.If you want to dig out effective data from vast amounts of data and find out the inextricable relationship between data,you need help with data mining..Cluster analysis is a method branch in data mining and has a wide range of applications.However,traditional clustering algorithms often have some drawbacks and need to be optimized and used in practical work.This paper first introduces the basic concepts of data mining and the more commonly used clustering algorithm K-means algorithm,which is a classical algorithm in cluster analysis.However,this algorithm is susceptible to the influence of the initial clustering center and does not necessarily obtain the optimal solution.It has certain limitations and is a local search technique.Aiming at the problems existing in clustering,the genetic algorithm is also introduced.The genetic algorithm is a biological type of evolutionary algorithm.According to Darwinian evolution theory and Mendelian genetic theory,natural selection is the best choice for the survival of the fittest.Although the genetic algorithm has its natural advantages,the same traditional genetic algorithm has premature convergence and is easily trapped in the local optimal value.Proposed the transformation of the existing genetic algorithm,the use of multi-group mechanism,the dynamic allocation of cross-mutation probability,the slow convergence speed in the simulated annealing algorithm,can jump out of the local optimal solution and other advantages,proposed a dynamic allocation of multi-group simulated annealing genetic algorithm.Combining this algorithm with the K-means algorithm in clustering,it has been used in clustering analysis and has played a good effect.Combining the optimized genetic algorithm with the clustering algorithm and experimenting with experimental data,the experimental data show that the optimized genetic algorithm and K-means algorithm use only the K-means algorithm and the standard genetic K-means algorithm.Must have certain advantages.The experimental results show that the dynamic allocation of multi-group simulated annealing K-means algorithm has a relatively significant improvement in the accuracy of the data.It shows that using the optimized genetic algorithm combined with the K-means algorithm will make the algorithm more efficient and reasonable,and the clustering effect is more excellent.It is suitable for cluster analysis. |