Abstract: Support Vector Machine（SVM），a kind of machine learning method，can efficiently solve the classification problem.A new classification-based feature selection algorithm is developed in this study.This algorithm is able to explore the best subset of features for classification from a group of either irrelevant or relevant features.Moreover，it can systematically prioritize all features based on degree of correlation between them and categories.And it finally is used to identify a set of combined-risk factors for type II diabetes in this study.A best subset of risk factors，consisting of waistline，waistline/hip-girth，diastolic blood pressure and age，is found for this disease.The sensitivity，specificity and accuracy of SVM classification under this subset are 0.866 6，0.642 0 and 0.701 4 respectively.In addition，a comparison between this algorithm and principal component analysis is also conducted.It turns out that the former is superior to the latter for the extraction of features.

Key words: SVM, feature selection, classification, typeⅡdiabetes

摘要： 支持向量机是一种基于结构风险最小化原理的分类技术，已逐渐引起国内外研究者的关注。提出了一种用于最佳特征子集选取的特征筛选算法，且实现了特征与分类识别相关性强度的排序，并通过使用该算法对Ⅱ型糖尿病判别与风险因素筛选，求证了该方法的可靠性和可行性。当以该算法提取的特征子集{腰围、腰围/臀围、舒张血压、年龄}作为输入向量时，敏感度、特异性、准确率最高，分别为0.866 6、0.642 0、0.701 4。同时，还将该算法与主成分分析法进行比较。实验表明，在特征提取方面该算法优于主成分分析法。因此，该算法对分类识别、风险因素筛选是一种有效的方法，为解决该类问题探索了一条有效途径。

关键词: 支持向量机, 特征提取, 分类识别, Ⅱ型糖尿病