【作者】 陈立萍； 乔元华；

【Author】 Chen Li-ping Qiao Yuan-hua(College of Applied Science, Beijing University of Technology, Beijing 100022)

【机构】 北京工业大学应用数理学院；

【摘要】 本文利用SVM方法对DNA序列样本进行判别分类,将编码区序列与非编码区序列作为两个总体,将输入空间映射到高维空间,以求在高维空间中问题线性可分或接近线性可分。再得到原空间的判别曲面,用所得到的判别曲面对编码区与非编码区有差异的特征序列进行划分。为了评估SVM方法分类的准确率,我们采用统计分析中的k-折交叉确认方法对样本数据进行了训练样本与测试,并将其结果与传统的判别分析法进行对照。实际处理结果表明在小样本情况下,SVM方法克服传统的判别分析方法对总体分布要求的局限性,分类效果优于传统的判别分析方法。更多还原

【Abstract】 SVM is a new classification technology based on the theory of statistic learning and idea of structural risk minimization.In many applications, SVM has been shown to provide higher performance than traditional learning machines and has been introduced as powerful tools for solving problems of classification. In this paper, we apply SVM into discriminant analysis of DNA sequences. Some concluding remarks and contrast to other traditional statistic discriminant analysis method are given in the end.更多还原