Multiclass Classification on High Dimension and Low Sample Size Data Using Genetic Programming

被引:5
作者
Wei, Tingyang [1 ]
Liu, Wei-Li [1 ]
Zhong, Jinghui [1 ]
Gong, Yue-Jiao [1 ]
机构
[1] South China Univ Technol, Sch Comp Sci & Engn, Guangzhou 510006, Peoples R China
关键词
Machine learning; Feature extraction; Gene expression; Programming; Genetic programming; Sociology; Statistics; gene expression programming; high dimension; classification; low sample size; ensemble learning; FEATURE-SELECTION; RULES;
D O I
10.1109/TETC.2020.3034495
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Multiclass classification is one of the most fundamental tasks in data mining. However, traditional data mining methods rely on the model assumption, they generally can suffer from the overfitting problem on high dimension and low sample size (HDLSS) data. Trying to address multiclass classification problems on HDLSS data from another perspective, we utilize Genetic Programming (GP), an intrinsic evolutionary classification algorithm that can implement feature construction automatically without model assumption. This article develops an ensemble-based genetic programming classification framework, the Sigmoid-based Ensemble Gene Expression Programming (SE-GEP). To relieve the problem of output conflict in GP-based multiclass classifiers, the proposed method employs a flexible probability representation with continuous relaxation to better integrate the output of all the binary classifiers, an effective data division strategy to further enhance the ensemble performance, and a novel sampling strategy to refine the existing GP-based binary classifier. The experiment results indicate that SE-GEP can attain better classification accuracy compared to other GP methods. Moreover, the comparison with other representative machine learning methods indicates that SE-GEP is a competitive method for multiclass classification in HDLSS data.
引用
收藏
页码:704 / 718
页数:15
相关论文
共 45 条
[21]  
Liu B, 2017, PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, P2287
[22]   Joint Binary Classifier Learning for ECOC-Based Multi-Class Classification [J].
Liu, Mingxia ;
Zhang, Daoqiang ;
Chen, Songcan ;
Xue, Hui .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2016, 38 (11) :2335-2341
[23]   ELDC: An Artificial Neural Network Based Energy-Efficient and Robust Routing Scheme for Pollution Monitoring in WSNs [J].
Mehmood, Amjad ;
Lv, Zhihan ;
Lloret, Jaime ;
Umar, Muhammad Muneer .
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING, 2020, 8 (01) :106-114
[24]  
Miller JF, 2000, LECT NOTES COMPUT SC, V1802, P121
[25]   Evolutionary constructive induction [J].
Muharram, M ;
Smith, GD .
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2005, 17 (11) :1518-1528
[26]   A novel approach to design classifiers using genetic programming [J].
Muni, DP ;
Pal, NR ;
Das, J .
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2004, 8 (02) :183-196
[27]   A Multiobjective Genetic Programming-Based Ensemble for Simultaneous Feature Selection and Classification [J].
Nag, Kaustuv ;
Pal, Nikhil R. .
IEEE TRANSACTIONS ON CYBERNETICS, 2016, 46 (02) :499-510
[28]   A Two-Layer Dimension Reduction and Two-Tier Classification Model for Anomaly-Based Intrusion Detection in IoT Backbone Networks [J].
Pajouh, Named Haddad ;
Javidan, Reza ;
Khayami, Raouf ;
Dehghantanha, Ali ;
Choo, Kim-kwang Raymond .
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING, 2019, 7 (02) :314-323
[29]  
Pedregosa F, 2011, J MACH LEARN RES, V12, P2825
[30]  
Rauss P.J., 2000, P GEN EV COMP C GECC, P726