| abstract |
The invention discloses a support vector machine parameter selection method based on a hybrid bat algorithm. Regularization parameters and RBF kernel parameters have a great influence on the learning performance and computational complexity of SVM. On the basis of analyzing the advantages and disadvantages of some classic parameter selection methods, the intelligent optimization algorithm is introduced to optimize its parameters. In view of the bat algorithm's advantages of parallelism, fast convergence speed, and strong robustness, the present invention first uses the bat algorithm to optimize the SVM parameters, and then introduces the crossover, selection and mutation operators of the differential evolution algorithm for the shortcoming of the bat algorithm's premature maturity , use the bat individual to further adjust the position according to the three operators in each iteration process, enhance the search ability of the algorithm, and avoid it from falling into the local optimal solution prematurely. Finally, the improved DEBA algorithm is used to optimize the selection of SVM parameters and achieve excellent results. |