摘要
在医疗诊断中 ,常根据病人的多项病理检测结果进行诊断。由于存在个体的差异和数据本身的噪声 ,所以要准确的诊断是困难的。支持向量机是在统计学习理论基础上发展而来的一种新的通用学习方法 ,具有很多独特的优点。该文介绍了支持向量机非线性分类算法 ,选取径向基核函数 ,构造了支持向量机非线性分类器 ,并将其应用于心脏病诊断。所用数据来自UCIbenchmark数据集。与其它方法相比 ,取得了较高的准确率。
Medical diagnosis carry out by data from multiple pathological examina tions. For the data are characterized by individual specificity, inherent noisy, it is difficult to accurately diagnose. Support Vector Machines (SVM) is a nove l powerful learning method developed on Statistical Learning Theory, and of many special advantages. SVM nonlinear classifiers algorithm is discussed in the pap er. Using radial basis function's kernels, SVM nonlinear classifier is employed to heart disease diagnoses based on UCI benchmark data set. Comparing other resu lt, high accuracy rate is obtained in the prediction. Application of SVM to dise ase diagnoses indicates SVM potential application in medical.
出处
《计算机仿真》
CSCD
2003年第2期69-70,63,共3页
Computer Simulation
