Research On Sparrow Search Algorithm Based On Fusion Of T Distribution And Tent Chaotic Mapping 2

Swarm intelligence optimization algorithm is a stochastic optimization algorithm.It is a meta-heuristic algorithm inspired by the self-optimization phenomenon of biological group behavior in nature.It mainly includes particle swarm optimization algorithm,ant colony optimization algorithm,artificial bee colony algorithm,firefly algorithm,gray wolf optimization algorithm and whale optimization algorithm,etc.The swarm intelligence optimization algorithm has the characteristics of simple structure,high stability,and support for parallel computing.It is suitable for large-scale and high-complexity scenarios.Therefore,swarm intelligence optimization algorithm has always been the focus of researchers.In the article,three improvement methods are proposed based on the sparrow search algorithm.Three sets of comparative verification experiments are designed using the benchmark function,and the improved sparrow search algorithms are applied to the classification of malicious software and malicious domain names.The main research contents are as follows:(1)Aiming at the shortcomings of the original sparrow search algorithm that it is easy to fall into local optimality,the convergence speed is slow and the convergence accuracy is not high,three improvement strategies are proposed.The first is to use t distribution to mutate the position of individual sparrows,so that individual sparrows can continuously explore outwards during the search process,which improves the ability of the algorithm to jump out of the local optimum;the second is to use the uniformity and ergodicity of Tent chaotic mapping,to generate the location information of the initial sparrow population,increase the diversity of the sparrow population,and improve the convergence speed and global search ability of the sparrow search algorithm;the third is to combine the t distribution and Tent chaotic mapping,and integrate the advantages of the two methods,and perturb the individuals trapped in the local optimum,further improve the ability of the algorithm to jump out of the local optimum.(2)Three sets of test experiments are designed,and seven high-dimensional single-objective test functions,five high-dimensional multi-objective test functions,and seven low-dimensional test functions are selected,which are mainly used to test the global search and local development capabilities of the improved sparrow search algorithms.Comparing experiments with the four swarm intelligent optimization algorithms,include the original sparrow search algorithm,firefly optimization algorithm,gray wolf optimization algorithm and whale optimization algorithm,the results show that the improved sparrow search algorithms have the advantages of fast convergence speed and high convergence accuracy.(3)The improved sparrow search algorithms are used to optimize the SVM classifier and BP neural network,and the models are applied to the classification of malicious software and malicious domain data sets.Comparing experiments with the four swarm intelligent optimization algorithms,include the original sparrow search algorithm,firefly optimization algorithm,gray wolf optimization algorithm and whale optimization algorithm,the results show that the improved sparrow search algorithms have a certain degree of optimization ability,which can improve the classification accuracy of malicious software and malicious domain names,and verify the effectiveness and practicability of the improved sparrow search algorithms.