基于融合Levy飞行策略与自适应变异因子优化QPSO-SVM算法

Optimization of QPSO-SVM algorithm based on fusion of Levy flight strategy and adaptive mutation factor

为了提高SVM分类精度与泛化能力,故提出一种基于融合Levy飞行策略与自适应变异因子优化QPSO-SVM算法.用量子粒子群算法(quantum particle swarm optimization,QPSO)对SVM进行惩罚因子和核函数参数优化,并针对QPSO算法出现的早熟收敛的问题,采用Levy飞行策略与自适应变异因子对其进行修正.实验仿真结果表明,与其他的智能优化算法相比,所提出的模型具有较高的分类性能和预测精度.

When classifying data,support vector machines(SVM)are widely used in hyperplane classification.However,if there are too many features of the data set in the application of SVM,the classification performance and fitting effect will be reduced,and the selection of internal parameters will also affect the classification performance and fitting effect of the SVM model.In order to improve the classification accuracy and generalization ability of SVM,an optimized QPSO-SVM algorithm based on Levy flight strategy and adaptive variation factor is proposed.The penalty factor and kernel function parameters of SVM were optimized by quantum particle swarm optimization(QPSO),and the Levy flight strategy and adaptive variation factor were used to correct the problem of precocious convergence of QPSO algorithm.Experimental simulation results show that compared with other intelligent optimization algorithms,the proposed model has high classification performance and prediction accuracy.