论文部分内容阅读
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 of SLE patients complicated with secondary Sjogren’s syndrome is not easy to diagnose and confirmed the subjectivity and so on, this paper proposed a computer-based support vector machine intelligent algorithm predictive diagnosis model. Firstly, 26 kinds of related diagnostic indexes of 141 patients in the material were preprocessed, making it a quantitative data suitable for SVM calculation. Secondly, using cross validation, grid search and improved Particle Swarm Optimization Support vector machine model of the penalty coefficient C and nuclear parameters g optimization selection, and use MATLAB software to draw the above three kinds of optimization methods were drawn support vector machine parameter model; the final comparison of selected SLE patients complicated by secondary drying Predictors of the highest predictive value of disease diagnosis. The results show that the self-optimization of SVM classification model based on improved Particle Swarm Optimization algorithm has the highest prediction accuracy.