An Improved Sparrow Search Algorithm for Optimizing Support Vector Machines

Support Vector Machine (SVM) is often used in regression and classification problems. However, SVM needs to find proper kernel function solve high-dimensional We propose an improved Sparrow Search Algorithm (ISSA-SVM) algorithm optimize the parameters. First, problem of slow convergence due lack ergodicity poor diversity initial population effectively overcome by using Sine chaotic map. Second, adaptive dynamic weight factors are induced not only balance global local search capabilities, but also accelerate speed sparrow algorithm. The simulation results 11 benchmark test functions show that ISSA has faster convergence, more accurate capability, easier jump out extremes than SSA, Gray Wolf Optimization (GWO) Whale (WOA). That indicates better robustness, stronger competitiveness. experimental on coal gangue dataset accuracy ISSA-SVM 7.09% 4.25% compared with SSA-SVM, respectively. Meanwhile, time for a single image frame reduced 20.15% 13.74%