| 基于双重正则化支持向量机的肿瘤基因选择 |
| |
| 引用本文： | 秦传东,刘三阳.基于双重正则化支持向量机的肿瘤基因选择[J].吉林大学学报(工学版),2013,43(1):192-197. |
| |
| 作者姓名： | 秦传东 刘三阳 |
| |
| 作者单位： | 1. 西安电子科技大学计算机学院,西安710071;北方民族大学信息与计算科学学院,银川750021
2. 西安电子科技大学理学院,西安,710071 |
| |
| 基金项目： | 国家自然科学基金项目(60974082);中央高校基本科研业务费专项项目(K50511700008) |
| |
| 摘 要： | 针对标准L2范数支持向量机和L1范数支持向量机在肿瘤基因分类分析中表现出的优缺点,在利用Bhattacharyya距离剔除部分对分类无关紧要特征基因,从而得到少数高相关至关重要特征基因的基础上,将一种双重正则化支持向量机应用到DNA微阵列分类中。用一种二次多项式损失函数把这种有约束的优化问题改变为无约束且可微的优化问题,这可以用BFGS算法来求解,通过对两种肿瘤特征基因数据集实验分析知,该算法对肿瘤特征基因分类具有较强的可行性和有效性。
|
| 关 键 词： | 计算机应用 基因表达谱 Bhattacharyya距离 双重正则化支持向量机 二次多项式损失函数 BFGS算法 |
Tumor gene selection based on double regularized support vector machine |
| |
| QIN Chuan-dong,LIU San-yang.Tumor gene selection based on double regularized support vector machine[J].Journal of Jilin University:Eng and Technol Ed,2013,43(1):192-197. |
| |
| Authors: | QIN Chuan-dong LIU San-yang |
| |
| Affiliation: | 1.School of Computer Seience and Technology,Xidian University,Xi’an 710071,China;2.School of Information and Computation Science,Beifang University of Nationalities,Yinchuan 750021,China;3.College of Mathematic Science,Xidian University,Xi’an 710071,China) |
| |
| Abstract: | According to the strengths and weaknesses of the L2-norm Support Vector Machine(SVM) and the L1-norm SVM in the classification analysis of cancer gene,a Doubly Regularized Support Vector Machine(DRSVM) is applied to the DNA microarray classification based on the Bhattacharyya distance,which is used to eliminate most of the unimportant genes and gain a few highly correlated important genes for classification.A quadratic polynomial loss function changes the constrained optimization into unconstrained and differentiable optimization,which can be computed by Brogden-Fltcher-Goldfarb-Shanno(BFGS) algorithm.Experiment results on two kinds of tumor gene data sets show that this method is effective and feasible. |
| |
| Keywords: | computer application gene expression profiles Bhattacharyya distance doubly regularized support vector machine quadratic polynomial loss function BFGS algorithm |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
