计算机与数字工程2019，Vol.47Issue(8)：1851-1855,5.DOI:10.3969/j.issn.1672-9722.2019.08.001
基于改进PSO-SVM算法的帕金森疾病诊断研究
Research of Parkinson's Disease Diagnosis Based on Improved PSO-SVM Algorithm
摘要
Abstract
The doctors easily give wrong judgments because of the etiology of Parkinson's disease unclear and diverse clinical manifestations. In this paper,a support vector machines algorithm based on improved particle swarm optimization(IMPSO-SVM)is proposed to improve the accuracy of recognition of Parkinson's disease. This algorithm assigns inertia weights and learning factors to different properties of particles to optimize the penalty coefficients and kernel functions of support vector machines. Finally,the pro?posed IMPSO-SVM algorithm is applied to the clinical data of Parkinson's disease. The experimental result shows that this algorithm has improved the prediction accuracy and the execution efficiency,compared with the support vector machine optimized by the tradi?tional particle swarm optimization(PSO-SVM)and the support vector machine optimized by genetic algorithm(GA-SVM). There?fore,this algorithm can be used as an effective method for assisting doctors to diagnose Parkinson's disease.关键词
改进粒子群算法/支持向量机/惯性权重/学习因子Key words
improved particle swarm optimization/support vector machines/inertia weight/learning factor分类
信息技术与安全科学引用本文复制引用
张琼,丁卫平,景炜,余利国..基于改进PSO-SVM算法的帕金森疾病诊断研究[J].计算机与数字工程,2019,47(8):1851-1855,5.基金项目
国家自然科学基金项目(编号:61300167) （编号:61300167）
江苏省自然科学基金项目(编号:BK20151274) （编号:BK20151274）
江苏省六大人才高峰项目(编号:XYDXXJS-048) （编号:XYDXXJS-048）
南通市应用基础研究项目(编号:GY12016014)资助. （编号:GY12016014）
