A Novel Improved Butterfly Optimization Algorithm Based on the Adaptive Inertia Weight and the Dynamic Switching Probability

Butterfly optimization algorithm (BOA) is widely applied in various complex problems because of its efficient performance. However, similar to other swarm intelligent optimization algorithms, the BOA also has the problem of easily falling into the local optimum. In order to solve the above problem, this paper proposes an improved BOA (IBOA) by balancing local search and global search. In the IBOA, the adaptive inertia weight and the dynamic switching probability are introduced, which enables the IBOA to significantly improve the exploration ability in the early stage and the exploitation ability in the later stage, so that the algorithm is able to effectively escape from the local optimum. The performance of the IBOA is verified on eight benchmark functions. Compared with the other four optimization algorithms, the IBOA has the strongest optimization ability. Finally, a new prediction model is proposed based on that the IBOA is used to determine the hyper-parameters of support vector machine (SVM), which is called the IBOA-SVM model. Prediction experiments are carried out in earthquake magnitude dataset and air quality index dataset to validate the performance of the IBOA-SVM. The comparative experimental results indicate that, in comparison to the SVM, GWO-SVM, PSO-SVM, and BOA-SVM model, the values of MSE and MAE obtained by the IBOA-SVM model are the smallest, so our proposed model has higher prediction accuracy.