Optimisation of Support Vector Machines based on Modified Chaos-mapping Grey Wolf Algorithm

The Grey Wolf algorithm exhibits shortcomings such as inadequate population initialization, slow convergence speed, and susceptibility to local optimization. This paper proposes improvements to the Grey Wolf algorithm, which is employ chaotic mapping and introduce a novel mechanism in the wolf pack update phase. Firstly, the chaotic mapping initialization of the grey wolf population increases population diversity and improve global search ability. Secondly, the combination of Lévy flight and whale algorithm improve the convergence speed of the algorithm and avoid falling into the local optimum. Finally, the performance of the algorithm was evaluated using a benchmark function, and the modified grey wolf algorithm was employed in the optimisation of diverse signals in the classification of support vector machines. This resulted in enhanced performance and higher accuracy to 95.5% in identifying the optimal solution.