计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (34): 246-248.

• 工程与应用 • 上一篇    

SVM在基因微阵列癌症数据分类中的应用

孟范静1,刘毅慧1,王洪国2,成金勇1   

  1. 1.山东轻工业学院 信息科学与技术学院,济南 250353
    2.山东省科学技术厅,济南 250011
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-12-01 发布日期:2007-12-01
  • 通讯作者: 孟范静

Cancer classification for microarray gene data based on SVM

MENG Fan-jing1,LIU Yi-hui1,WANG Hong-guo2,CHENG Jin-yong1   

  1. 1.Department of Computer Science and Information Technology,Shandong Institute of Light Industry,Ji’nan 250353,China
    2.Department of Science & Technology of Shandong Province,Ji’nan 250011,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-12-01 Published:2007-12-01
  • Contact: MENG Fan-jing

摘要: 在总结二分类支持向量机应用的基础上,提出了利用t-验证方法和Wilcoxon验证方法进行特征选取,以支持向量机(SVM)为分类器,针对基因微阵列癌症数据进行分析的新方法,通过对白血病数据集和结肠癌数据集的分类实验,证明提出的方法不但识别率高,而且需要选取的特征子集小,分类速度快,提高了分类的准确性与分类速度。

关键词: 微阵列数据, 支持向量机, 癌症数据分类, 特征选取

Abstract: This paper analyzes microarray data based on support vector machine.T-test and Wilcoxon methods are used for feature selection and dimensionality reduction.Support vector machine is employed for the classification of microarray data.Experiments are performed on leukemia and colon sets.Good performance is achieved and this method also reduces computation time with small feature subset.

Key words: microarray data, Support Vector Machine(SVM), cancer classification, feature selection