Research on SVM Algorithm Based on Density and Particle Swarm Optimization
TANG Ying
YAN Renwu
Abstract：In the training process of SVM algorithm,the training sample set is too large,resulting in redundant samples and noise in the training process,which will affect the partition hyperplane.Therefore,a density based clipping method is proposed to clip the training samples,remove the noise and redundant samples of the sample set,optimize the partition of the classification hy-perplane and improve the classification accuracy.The penalty factor C and kernel function g in SVM algorithm have a great impact on the classification performance.The SVM algorithm is improved based on particle swarm optimization algorithm,and the penalty factor C and kernel parameter g are optimized by particle swarm optimization algorithm.Using the sample set to demonstrate the im-proved algorithm,this method effectively improves the SVM classification performance.
Keywords：SVM algorithmdensityparticle swarm optimizationmachine learning
Publication Date：2023-10-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 2257-2262 )