Table of Contents Author Guidelines Submit a Manuscript
Mathematical Problems in Engineering
Volume 2014 (2014), Article ID 915058, 16 pages
http://dx.doi.org/10.1155/2014/915058
Research Article

Component Thermodynamical Selection Based Gene Expression Programming for Function Finding

1School of Science, JiangXi University of Science and Technology, Ganzhou 341000, China
2State Key Laboratory of Software Engineering, Wuhan University, Wuhan 430072, China
3Computer School, Wuhan University, Wuhan 430072, China
4State-Owned Assets Supervision and Administration of Jiangxi Province, Nanchang 330006, China
5College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
6School of Information Engineering, Shijiazhuang University of Economics, Shijiazhuang 050031, China

Received 8 June 2013; Accepted 8 December 2013; Published 16 January 2014

Academic Editor: Wei-Chiang Hong

Copyright © 2014 Zhaolu Guo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Gene expression programming (GEP), improved genetic programming (GP), has become a popular tool for data mining. However, like other evolutionary algorithms, it tends to suffer from premature convergence and slow convergence rate when solving complex problems. In this paper, we propose an enhanced GEP algorithm, called CTSGEP, which is inspired by the principle of minimal free energy in thermodynamics. In CTSGEP, it employs a component thermodynamical selection (CTS) operator to quantitatively keep a balance between the selective pressure and the population diversity during the evolution process. Experiments are conducted on several benchmark datasets from the UCI machine learning repository. The results show that the performance of CTSGEP is better than the conventional GEP and some GEP variations.