The SVM Classifier Based on the Modified Particle Swarm Optimization
This work addresses the challenge of efficiently tuning SVM parameters for data classification, but it appears incremental as it builds on existing particle swarm optimization methods.
The authors tackled the problem of developing an SVM classifier by using a modified particle swarm optimization algorithm to simultaneously search for kernel function type, parameters, and regularization, resulting in high classification quality and reduced time expenditures, as confirmed by experimental studies.
The problem of development of the SVM classifier based on the modified particle swarm optimization has been considered. This algorithm carries out the simultaneous search of the kernel function type, values of the kernel function parameters and value of the regularization parameter for the SVM classifier. Such SVM classifier provides the high quality of data classification. The idea of particles' «regeneration» is put on the basis of the modified particle swarm optimization algorithm. At the realization of this idea, some particles change their kernel function type to the one which corresponds to the particle with the best value of the classification accuracy. The offered particle swarm optimization algorithm allows reducing the time expenditures for development of the SVM classifier. The results of experimental studies confirm the efficiency of this algorithm.