Parameters optimisation of support vector machine using modified grasshopper optimisation algorithm-based Levy-flight method

摘要

Grasshopper optimisation algorithm (GOA) is one of the most recent meta-heuristic optimisation algorithms. It was first developed by Saremi et al. in 2017. Although, GOA has shown good performance, it still has demerits with respect to low precision, slow convergence and easily stuck at local minima. This paper presents a modified version of GOA based on Lèvy-flight method, called as (LevyGOA). The proposed LevyGOA is proved to provide a better trade-off between exploitation and exploration, which makes LevyGOA faster and more robust than GOA. LevyGOA is further compared with other meta-heuristic optimisation algorithms and the basic GOA for solving two optimisation problems. These problems are global optimisation problem and parameters optimisation of SVM, where CEC 2005 and CEC 2017 global benchmark functions and six well-known benchmark datasets are used. The experimental results show that LevyGOA can significantly improve the performance of GOA. The results demonstrate that LevyGOA outperforms the other algorithms on a majority of the benchmark functions and benchmark datasets.