Research on Parkinson's Disease Diagnosis Based on Improved Particle Swarm Optimization Support Vector Machine Algorithm

In order to solve the problem of Parkinson's disease(PD) is unclear and variety of clinical manifestations of PD, which easy to cause doctors to misjudge. In this paper, a support vector machine algorithm(SVM) based on improved particle swarm optimization(PSO) is proposed to diagnose of PD, which is used to improve the recognition accuracy of PD. By improving the PSO algorithm, the algorithm dynamically assigns inertia weights and learning factors, which are trained with different performances of particle, and optimizes the penalty factor and kernel function of the support vector machine to obtain the global optimal solution to improve the learning ability and performance of the support vector machine model. Finally, the algorithm proposed in this paper is applied to the data of clinical manifestations of PD, and the experiments show that the algorithm is better than the classical support vector machine algorithm based on particle swarm optimization(PSO-SVM) and the support vector machine algorithm based on genetic optimization algorithm(GA-SVM). The execution efficiency has been improved. Therefore, the algorithm can be used as an effective method to assist doctors in diagnosing of PD.