| 一种基于支持向量机的自适应肿瘤分类检测算法 |
| |
| 引用本文： | 黄伟,尹京苑.一种基于支持向量机的自适应肿瘤分类检测算法[J].生物信息学,2009,7(4):243-247. |
| |
| 作者姓名： | 黄伟 尹京苑 |
| |
| 作者单位： | 上海大学生命科学学院生物信息学中心,上海,200444 |
| |
| 摘 要： | 根据肿瘤分类检测模型的特点，提出了一种新的算法，该算法结合使用了基因选择和数据抽取的有效方法，并在此基础上使用支持向量机对基因表达数据进行分类或者检测。其中乳腺癌的分类交叉验证结果由88．46％提高到100．0％，急性白血病的也由71．05％提高至100．0％。实验结果说明了这一方法的有效性，为在大量的基因表达数据中提高检测癌症的准确性提出了一种比较通用的方法。
|
| 关 键 词： | 非参数检验 支持向量机 肿瘤分类检测 自适应算法 |
An Self-adaptive Cancer Classification Algorithm Based on Support Vector Machine |
| |
| HUANG Wei,YIN Jing-yuan.An Self-adaptive Cancer Classification Algorithm Based on Support Vector Machine[J].China Journal of Bioinformation,2009,7(4):243-247. |
| |
| Authors: | HUANG Wei YIN Jing-yuan |
| |
| Institution: | ( Bioinformatics Center, College of Life Science, Shanghai University, Shanghai 200444, China) |
| |
| Abstract: | A new algorithm which utilizes the effective methods in combining gene selection and data extraction is proposed according to the features of the tumor classification and detection model. And support vector machine is used to classify or detect tumor. The cross validation accuracy of breast cancer improves from 88.46% to 100.0% while leukaemia's from 71.05% to 100.0%. This experiment result shows that the algorithm is a more general means which can be used in vast gene expression profile data. |
| |
| Keywords: | Non-parametric Test Support Vector Machine Tumor Identification and Classification Self-adaptive algorithm |
| 本文献已被 维普 万方数据 等数据库收录！ |
