[1]李钧涛,贾英民.PCD型自适应弹性网络在微阵列分类中的应用[J].智能系统学报,2010,5(03):227-232.
 LI Jun-tao,JIA Ying-min.Applying a PCD adaptive elastic net in microarray classification[J].CAAI Transactions on Intelligent Systems,2010,5(03):227-232.
点击复制

PCD型自适应弹性网络在微阵列分类中的应用(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第5卷
期数:
2010年03期
页码:
227-232
栏目:
出版日期:
2010-06-25

文章信息/Info

Title:
Applying a PCD adaptive elastic net in microarray classification
文章编号:
1673-4785(2010)03-0227-06
作者:
李钧涛贾英民
北京航空航天大学 第七研究室,北京 100191
Author(s):
LI Jun-tao JIA Ying-min
The Seventh Research Division, Beihang University, Beijing 100191, China
关键词:
癌症分类基因选择弹性网络顺向坐标下降算法(PCD算法)微阵列分类
Keywords:
cancer classification gene selection elastic net pathwise coordinate descent algorithm microarray classification
分类号:
TP273
文献标志码:
A
摘要:
针对癌症分类中的重要基因选择问题,提出了一种基于顺向坐标下降算法的自适应弹性网络.该自适应弹性网络通过引入数据驱动权重,在构建分类器的同时能自适应地成群选择基因,从而产生了一个稀疏的学习模型,增强了可解释性.此外,通过引入惩罚因子,顺向坐标下降算法被改进并有效地用于求解该自适应弹性网络.急性白血病分类实验结果验证了所提方法的有效性.
Abstract:
An adaptive elastic net was proposed, based on a pathwise coordinate descent (PCD) algorithm, to select genes important for cancer classification. By introducing datadriven weights, the proposed adaptive elastic net can adaptively select genes in groups in the process of building classifiers. It thus produces a sparse learning model with enhanced interpretability. Furthermore, by introducing penalty factors, the pathwise coordinate descent algorithm was improved, solving the adaptive elastic net more efficiently. Experimental results from leukemia classification verified the proposed method. 

参考文献/References:

[1]GOLUB T R, SLONIM D K, TAMAYO P, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring[J]. Science, 1999, 286(5439): 531536.
 [2]GUYON I, WESTON J, BARNHILL S, VAPNIK V. Gene selection for cancer classification using support vector machines[J]. Machine Learning, 2002, 46(1): 389422.
[3]ZHU J, ROSSET S, HASTIE T, TIBSHIRANI R. 1norm support vector machines[J]. Advances in Neural Information Processing Systems, 2004, 16(1): 4956.
[4]SHEVADE S K, KEERTHI S S. A simple and efficient algorithm for gene selection using sparse logistic regression[J]. Bioinformatics, 2006, 19(17): 22462253.
[5]CAWLEY G C, TALBOT N L C. Gene selection in cancer classification using sparse logistic regression with Bayesian regularization[J]. Bioinformatics, 2006, 22(19): 23482355.
[6]TIBSHIRANI R. Regression shrinkage and selection via the lasso[J]. Journal of the Royal Statistical Society: Series B, 1996, 58(1): 267288
[7]ZOU Hui, HASTIE T. Regularization and variable selection via the elastic net[J]. Journal of the Royal Statistical Society: Series B, 2005, 67: 301320.
[8]WANG Li, ZHU Ji, ZOU Hui. The doubly regularized support vector machine[J]. Statistica Sinica, 2006, 16(2): 589615.
[9]李钧涛,贾英民. 用于微阵列分类的Huberized多类支持向量机[J]. 自动化学报, 2010, 36(3): 399405.
LI Juntao, JIA Yingmin. Huberized multiclass support vector machine for microarray classification[J]. Acta Automatica Sinica, 2010, 36(3): 399405.
[10]李钧涛,贾英民. 用于癌症分类与基因选择的一种改进的弹性网络[J]. 自动化学报, 2010, 36(7): 976981.
 LI Juntao, JIA Yingmin. An improved elastic net for cancer classification and gene selection[J]. Acta Automatica Sinica, 2010, 36(7): 976981.
[11]LI Juntao, JIA Yingmin, DU Junping, YU Fashan. A new support vector machine for microarray classification and adaptive gene selection[C]//2009 American Control Conference. St. Louis, USA: 54105415.
[12]EGAL M, DAHLQUIST K, CONKLIN B. Regression approaches for microarray data analysis[J]. Journal of Computational Biology, 2003, 10(6): 961980.
[13]VAPNIK V. The nature of statistical learning theory[M]. New York: Springer, 1995: 160.
[14]EFRON B, HASTIE T, JOHNSTON I, TIBSHIRANI R. Least angle regression[J]. Annals of Statistics, 2004, 32(2): 407499.
[15]陈晓峰,王士同,曹苏群. 半监督多标记学习的基因功能分析[J]. 智能系统学报, 2008, 3(1): 8390.
CHEN Xiaofeng, WANG Shitong, CAO Suqun. Gene function analysis of semisupervised multilabel learning[J]. CAAI Transactions on Intelligent Systems, 2008, 3(1): 8390.
[16]FRIEDMAN J, HASTIE T, TIBSHIRANI R. Regularization paths for generalized linear models via coordinate descent[R]. Palo Alto, USA: Standford University, 2008.
[17]FRIEDMAN J, HASTIE T, HFLING H, TIBSHIRANI R. Pathwise coordinate optimization[J]. Annals of Applied Statistics, 2007, 1(2): 302332.
 [18]GHOSH S. Adaptive elastic net: an improvement of elastic net to achieve oracle properties: IUPUI tech report No. pr0701[R]. Indianapolis, USA: Department of Mathematical Sciences, Indiana UniversityPurdue University, 2007.
[19]ZOU H, ZHANG H H. On the adaptive elastic net with a diverging number of parameters[J]. Annals of Statistics, 2009, 37(4): 17331751.

备注/Memo

备注/Memo:
收稿日期:2009-12-14.
基金项目:国家自然科学基金资助项目(60727002, 60774003, 60850004);国家“973”计划资助项目(2005CB321902);国防基础研究资助项目(A2120061303).
通信作者:李钧涛.     E-mail:juntaolimail@yahoo.com.cn.
作者简介:
李钧涛,男,1978年生,讲师、博士.主要研究方向为智能控制、统计学习及其在生物信息学中的应用. 
 贾英民,男,1958年生,教授、博士生导师,教育部“长江学者”特聘教授,中国科学院系统控制重点实验室学术委员会委员,中国人工智能学会智能空天系统专业委员会主任,中国自动化学会控制理论专业委员会副主任,中国航空学会控制理论与应用专业委员会副主任.主要研究方向为鲁棒控制、自适应控制、智能控制及其在车辆系统和工业过程中的应用.承担国家“973”计划、“863”计划,国家自然科学基金重点项目、科学仪器专项,面上项目,国防基础科研项目,教育部高校博士点基金等20余项.国家杰出青年科学基金获得者,国家“百千万人才工程”第一、二层次人选.发表学术论文120余篇,出版专著1部,申请专利10余项.
更新日期/Last Update: 2010-07-14