Genetic programming for feature construction and selection in classification on high-dimensional data

被引:136
作者
Binh Tran [1 ]
Xue, Bing [1 ]
Zhang, Mengjie [1 ]
机构
[1] Victoria Univ Wellington, Evolutionary Computat Res Grp, POB 600, Wellington 6140, New Zealand
关键词
Genetic programming; Feature construction; Feature selection; Classification; High-dimensional data; ALGORITHM; OPTIMIZATION; CLASSIFIERS;
D O I
10.1007/s12293-015-0173-y
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Classification on high-dimensional data with thousands to tens of thousands of dimensions is a challenging task due to the high dimensionality and the quality of the feature set. The problem can be addressed by using feature selection to choose only informative features or feature construction to create new high-level features. Genetic programming (GP) using a tree-based representation can be used for both feature construction and implicit feature selection. This work presents a comprehensive study to investigate the use of GP for feature construction and selection on high-dimensional classification problems. Different combinations of the constructed and/or selected features are tested and compared on seven high-dimensional gene expression problems, and different classification algorithms are used to evaluate their performance. The results show that the constructed and/or selected feature sets can significantly reduce the dimensionality and maintain or even increase the classification accuracy in most cases. The cases with overfitting occurred are analysed via the distribution of features. Further analysis is also performed to show why the constructed feature can achieve promising classification performance.
引用
收藏
页码:3 / 15
页数:13
相关论文
共 33 条
[1]  
Ahmed Soha, 2012, AI 2012: Advances in Artificial Intelligence. 25th Australasian Conference. Proceedings, P266, DOI 10.1007/978-3-642-35101-3_23
[2]   Multiple Feature Construction for Effective Biomarker Identification and Classification using Genetic Programming [J].
Ahmed, Soha ;
Zhang, Mengjie ;
Peng, Lifeng ;
Xue, Bing .
GECCO'14: PROCEEDINGS OF THE 2014 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, 2014, :249-256
[3]  
Ahmed S, 2013, 2013 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), P584
[4]   Selection bias in gene extraction on the basis of microarray gene-expression data [J].
Ambroise, C ;
McLachlan, GJ .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2002, 99 (10) :6562-6566
[5]  
[Anonymous], 2010, EVOLUTIONARY FEATURE
[6]  
Banzhaf Wolfgang, 1998, Genetic programming: an introduction on the automatic evolution of computer programs and its applications
[7]   Reusing Genetic Programming for Ensemble Selection in Classification of Unbalanced Data [J].
Bhowan, Urvesh ;
Johnston, Mark ;
Zhang, Mengjie ;
Yao, Xin .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2014, 18 (06) :893-908
[8]   A survey on feature selection methods [J].
Chandrashekar, Girish ;
Sahin, Ferat .
COMPUTERS & ELECTRICAL ENGINEERING, 2014, 40 (01) :16-28
[9]   A GA-based feature selection approach with an application to handwritten character recognition [J].
De Stefano, C. ;
Fontanella, F. ;
Marrocco, C. ;
di Freca, A. Scotto .
PATTERN RECOGNITION LETTERS, 2014, 35 :130-141
[10]   Minimum redundancy feature selection from microarray gene expression data [J].
Ding, C ;
Peng, HC .
PROCEEDINGS OF THE 2003 IEEE BIOINFORMATICS CONFERENCE, 2003, :523-528