SVM分类器在继发性干燥综合征诊断中的应用
|
摘要
本文为解决SLE患者并发继发性干燥综合征不易诊断及确诊主观性较强等问题,提出了一种可供计算机学习的支持向量机智能算法预测诊断模型。首先对材料中141名患者的26种相关诊断指标进行数据预处理,使之成为能够适合支持向量机计算的量化数据;其次运用交叉验证法、网格搜索法、改进的粒子群优化算法分别对支持向量机模型中的惩罚系数C与核参数g进行优化选择,并利用MATLAB软件分别画出以上3种优化方式得出的支持向量机参数模型;最终对比选出对SLE患者并发继发性干燥综合征疾病诊断预测度最高的预测模型。结果表明,基于改进的粒子群算法优化的支持向量机分类模型参数的自优化,对该疾病预测诊断精度最高。
In order to solve the problem that SLE patients complicated with secondary dryness syndrome is not easy to be diagnosed and diagnosed with strong subjectivity,a support vector machine intelligent algorithm for computer learning is proposed in this paper.First of materials in 141 patients with 26 kinds of diagnosis index of data pre processing,make it suitable for calculation of SVM is the quantitative data;followed by the use of cross validation method,grid search method,improved particle swarm optimization algorithm respectively to support vector machine model in the penalty factor C and kernel parameter g were optimized,and the use of MATLAB software were painted in the above three kinds of optimization methods that support vector machine parameter model;comparison choose of SLE patients with secondary Sjogren syndrome disease diagnosing and predicting the highest degree of prediction model.The results show that the algorithm based on improved particle swarm optimization algorithm to optimize the parameters of the support vector machine classification model,the highest accuracy of the prediction of the disease.
In order to solve the problem that SLE patients complicated with secondary dryness syndrome is not easy to be diagnosed and diagnosed with strong subjectivity,a support vector machine intelligent algorithm for computer learning is proposed in this paper.First of materials in 141 patients with 26 kinds of diagnosis index of data pre processing,make it suitable for calculation of SVM is the quantitative data;followed by the use of cross validation method,grid search method,improved particle swarm optimization algorithm respectively to support vector machine model in the penalty factor C and kernel parameter g were optimized,and the use of MATLAB software were painted in the above three kinds of optimization methods that support vector machine parameter model;comparison choose of SLE patients with secondary Sjogren syndrome disease diagnosing and predicting the highest degree of prediction model.The results show that the algorithm based on improved particle swarm optimization algorithm to optimize the parameters of the support vector machine classification model,the highest accuracy of the prediction of the disease.
引文
[1]李琳,张晓龙.基于RBF核的SVM学习算法的优化计算[J].计算机工程与应用,2006(29):190-204.
[2]汪海燕,黎建辉,杨风雷.支持向量机理论及算法研究综述[J].计算机应用研究,2014,31(5):1281-1286.
[3]FEI B,LIU J.Binary tree of SVM:a new fast muhiclass training and classification algorithm[J].IEEE Trans.On Neural Networks,2006,17(3):696-704.
[4]陈鲤江,景程,吴姚鑫,等.数学表达式的归一化方法研究[J].浙江工业大学学报,2012,40(2):229-236.
[5]周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用[J].电子测量与仪器学报,2014,28(3):240-245.
[6]张公让,万飞.基于网格搜索的SVM在入侵检测中的应用[J].计算机技术与发展,2016(1):97-100.
[7]PLATT J.Fast training of support vector machines using sequential minimal optimization[C]//Advances in Kernel MethodsSupport Vector Learning.Cambridge,MA:MIT Press,1998.
[8]王兴玲,李占斌.基于网格搜索的支持向量机核函数参数的确定[J].中国海洋大学学报,2005,35(5):859-862.
[9]杨柳松,何光宇.基于改进粒子群优化的SVM故障诊断方法[J].计算机工程,2013,39(3):187-196.
[10]张利彪,周春光,马铭,等.基于粒子群算法求解多目标优化问题[J].计算机研究与发展,2004,41(7):1286-1291.
[11]Shi Y,Eberhart R C.A Modified Particle Swarm Optimizer[C]//Proc.of IEEE International Conference on Evolutionary Computation.Piscataway,USA:IEEE Service Center,1998:69-73.
[2]汪海燕,黎建辉,杨风雷.支持向量机理论及算法研究综述[J].计算机应用研究,2014,31(5):1281-1286.
[3]FEI B,LIU J.Binary tree of SVM:a new fast muhiclass training and classification algorithm[J].IEEE Trans.On Neural Networks,2006,17(3):696-704.
[4]陈鲤江,景程,吴姚鑫,等.数学表达式的归一化方法研究[J].浙江工业大学学报,2012,40(2):229-236.
[5]周绍磊,廖剑,史贤俊.RBF-SVM的核参数选择方法及其在故障诊断中的应用[J].电子测量与仪器学报,2014,28(3):240-245.
[6]张公让,万飞.基于网格搜索的SVM在入侵检测中的应用[J].计算机技术与发展,2016(1):97-100.
[7]PLATT J.Fast training of support vector machines using sequential minimal optimization[C]//Advances in Kernel MethodsSupport Vector Learning.Cambridge,MA:MIT Press,1998.
[8]王兴玲,李占斌.基于网格搜索的支持向量机核函数参数的确定[J].中国海洋大学学报,2005,35(5):859-862.
[9]杨柳松,何光宇.基于改进粒子群优化的SVM故障诊断方法[J].计算机工程,2013,39(3):187-196.
[10]张利彪,周春光,马铭,等.基于粒子群算法求解多目标优化问题[J].计算机研究与发展,2004,41(7):1286-1291.
[11]Shi Y,Eberhart R C.A Modified Particle Swarm Optimizer[C]//Proc.of IEEE International Conference on Evolutionary Computation.Piscataway,USA:IEEE Service Center,1998:69-73.