[1]滕旭阳,董红斌,孙静.面向特征选择问题的协同演化方法[J].智能系统学报,2017,12(01):24-31.[doi:10.11992/tis.201611029]
 TENG Xuyang,DONG Hongbin,SUN Jing.Co-evolutionary algorithm for feature selection[J].CAAI Transactions on Intelligent Systems,2017,12(01):24-31.[doi:10.11992/tis.201611029]
点击复制

面向特征选择问题的协同演化方法(/HTML)
分享到:

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

卷:
第12卷
期数:
2017年01期
页码:
24-31
栏目:
出版日期:
2017-02-25

文章信息/Info

Title:
Co-evolutionary algorithm for feature selection
作者:
滕旭阳 董红斌 孙静
哈尔滨工程大学 计算机科学与技术学院, 黑龙江 哈尔滨 150001
Author(s):
TENG Xuyang DONG Hongbin SUN Jing
College of Computer Science and Technology, Harbin Engineering University, Harbin 150001, China
关键词:
特征选择遗传算法粒子群优化协同演化比特率交叉
Keywords:
feature selectiongenetic algorithm (GA)particle swarm optimization (PSO)co-evolutionbit rate cross
分类号:
TP301
DOI:
10.11992/tis.201611029
摘要:
特征选择技术是机器学习和数据挖掘任务的关键预处理技术。传统贪婪式特征选择方法仅考虑本轮最佳特征,从而导致获取的特征子集仅为局部最优,无法获得最优或者近似最优的特征集合。进化搜索方式则有效地对特征空间进行搜索,然而不同的进化算法在搜索过程中存在自身的局限。本文吸取遗传算法(GA)和粒子群优化算法(PSO)的进化优势,以信息熵度量为评价,通过协同演化的方式获取最终特征子集。并提出适用于特征选择问题特有的比特率交叉算子和信息交换策略。实验结果显示,遗传算法和粒子群协同进化(GA-PSO)在进化搜索特征子集的能力和具体分类学习任务上都优于单独的演化搜索方式。进化搜索提供的组合判断能力优于贪婪式特征选择方法。
Abstract:
Feature selection is a key preprocessing technology of machine learning and data mining. The traditional greed type of feature selection methods only considers the best feature of the current round, thereby leading to the feature subset that is only locally optimal. Realizing an optimal or nearly optimal feature set is difficult. Evolutionary search means can effectively search for a feature space, but different evolutionary algorithms have their own limitations in search processes. The evolutionary advantages of genetic algorithms (GA) and particle swarm optimization (PSO) are absorbed in this study. The final feature subset is obtained by co-evolution, with the information entropy measure as an assessment function. A specific bit rate cross operator and an information exchange strategy applicable for a feature selection problem are proposed. The experimental results show that the co-evolutionary method (GA-PSO) is superior to the single evolutionary search method in the search ability of the feature subsets and classification learning. In conclusion, the ability of combined evaluation, which is provided by an evolutionary search, is better than that of the traditional greedy feature selection method.

参考文献/References:

[1] DASH M, LIU H. Feature selection for classification[J]. Intelligent data analysis, 1997, 1(1/2/3/4): 131-156.
[2] GUYON I, ELISSEEFF A. An introduction to variable and feature selection[J]. The journal of machine learning research, 2002, 3(6): 1157-1182.
[3] ZHAO Zheng, MORSTATTER F, SHARMA S, et al. Advancing feature selection research. ASU feature selection repository[R]. Phoenix: School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, Tempe, 2010.
[4] BATTITI R. Using mutual information for selecting features in supervised neural net learning[J]. IEEE transactions on neural networks, 1994, 5(4): 537-550.
[5] YANG Yiming, PEDEREN J O. A comparative study on feature selection in text categorization[C]//Proceedings of the 14th International Conference on Machine Learning. San Francisco, CA, USA 1997: 412-420.
[6] 周志华. 机器学习[M]. 北京: 清华大学出版社, 2016: 247-266.
[7] XUE Bing, ZHANG Mengjie, BROWNE W N, et al. A survey on evolutionary computation approaches to feature selection[J]. IEEE transactions on evolutionary computation, 2016, 20(4): 606-626.
[8] PENG Hanchuan, LONG Fuhui, DING C. Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy[J]. IEEE transactions on pattern analysis and machine intelligence, 2005, 27(8): 1226-1238.
[9] UNLER A, MURAT A, CHINNAM R B. Mr2PSO: a maximum relevance minimum redundancy feature selection method based on swarm intelligence for support vector machine classification[J]. Information sciences, 2011, 181(20): 4625-4641.
[10] CERVANTE L, XUE Bing, ZHANG Mengjie, et al. Binary particle swarm optimisation for feature selection: a filter based approach[C]//Proceedings of 2012 IEEE Congress on Evolutionary Computation. Piscataway. Brisbane, Australia, 2012: 1-8.
[11] DONG Hongbin, TENG Xuyang, ZHOU Yang, et al. Feature subset selection using dynamic mixed strategy[C]//Proceedings of 2015 IEEE Congress on Evolutionary Computation. Sendai, Japan, 2015: 672-679.
[12] NEMATI S, BASIRI M E, GHASEM-AGHAEE N, et al. A novel ACO-GA hybrid algorithm for feature selection in protein function prediction[J]. Expert systems with applications, 2009, 36(10): 12086-12094.
[13] KENNEDY J, EBERHART R. Particle swarm optimization[C]//Proceedings of 1995 IEEE International Conference on Neural Networks. Perth, Australia, 1995: 1942-1948.
[14] KENNEDY J, EBERHART R. A discrete binary version of the particle swarm algorithm[C]//Proceedings of 1997 IEEE International Systems, Man, and Cybernetics. Orlando, USA, 1997: 4104-4108.
[15] 李书全, 孙雪, 孙德辉, 等. 遗传算法中的交叉算子的述评[J]. 计算机工程与应用, 2012, 48(1): 36-39. LI Shuquan, SUN Xue, SUN Dehui, et al. Summary of crossover operator of genetic algorithm[J]. Computer engineering and applications, 2012, 48(1): 36-39.

相似文献/References:

[1]周本达,陈明华.随机化均匀设计混合遗传算法求解图的二划分问题[J].智能系统学报,2009,4(01):91.
 ZHOU Ben-da,CHEN Ming-hua.Solving the 2-way graph partitioning problem using a genetic algorithm based on randomized uniform design[J].CAAI Transactions on Intelligent Systems,2009,4(01):91.
[2]孙正兴,张尧烨,李 彬.基于线性规划分类器的相关反馈技术[J].智能系统学报,2007,2(03):34.
 SUN Zheng-xing,ZHANG Yao-ye,LI Bin.Applying relevance feedback with a linear programming classifier[J].CAAI Transactions on Intelligent Systems,2007,2(01):34.
[3]康 琦,汪 镭,刘小莉,等.基于群体智能框架理念的遗传算法总体模式描述[J].智能系统学报,2007,2(05):42.
 KANG Qi,WANG Lei,LIU Xiao-li,et al.General mode description genetic algorithms based on a framework of swarm intelligence[J].CAAI Transactions on Intelligent Systems,2007,2(01):42.
[4]马 炫,张亚龙.基于遗传算法的大规模矩形件优化排样[J].智能系统学报,2007,2(05):48.
 MA Xuan,ZHANG Ya-long.A genetic algorithm for the layout of large scale rectang ular parts[J].CAAI Transactions on Intelligent Systems,2007,2(01):48.
[5]徐 雄.人工情感的进化控制系统实现[J].智能系统学报,2008,3(02):135.
 XU Xiong.Implementation of an evolutionary control system based on artificial emotion[J].CAAI Transactions on Intelligent Systems,2008,3(01):135.
[6]刘 胜,李高云,孙天英.一种基于种群多样度的实数编码并行遗传算法[J].智能系统学报,2008,3(05):423.
 L IU Sheng,L I Gao-yun,SUN Tian-ying.A real coding parallel genetic algorithm based on diversity of population[J].CAAI Transactions on Intelligent Systems,2008,3(01):423.
[7]张 涛,费树岷,李晓东.基于GARBF神经网络及边界不变特征的车辆识别[J].智能系统学报,2009,4(03):278.
 ZHANG Tao,FEI Shu-min,LI Xiao-dong.Vehicle recognition using boundary invariants and a genetic algorithm trained radial basis function neural network[J].CAAI Transactions on Intelligent Systems,2009,4(01):278.
[8]秦世引,高书征.面向救援任务的地面移动机器人路径规划[J].智能系统学报,2009,4(05):414.[doi:10.3969/j.issn.1673-4785.2009.05.005]
 QIN Shi-yin,GAO Shu-zhen.Path planning for mobile rescue robots in disaster areas with complex environments[J].CAAI Transactions on Intelligent Systems,2009,4(01):414.[doi:10.3969/j.issn.1673-4785.2009.05.005]
[9]张志飞,苗夺谦.基于粗糙集的文本分类特征选择算法[J].智能系统学报,2009,4(05):453.[doi:10.3969/j.issn.1673-4785.2009.05.011]
 ZHANG Zhi-fei,MIAO Duo-qian.Feature selection for text categorization based on rough set[J].CAAI Transactions on Intelligent Systems,2009,4(01):453.[doi:10.3969/j.issn.1673-4785.2009.05.011]
[10]周树德,孙增圻.遗传算法中的联结关系[J].智能系统学报,2009,4(06):483.[doi:10.3969/j.issn.1673-4785.2009.06.003]
 ZHOU Shu-de,SUN Zeng-qi.Linkage in genetic algorithms[J].CAAI Transactions on Intelligent Systems,2009,4(01):483.[doi:10.3969/j.issn.1673-4785.2009.06.003]

备注/Memo

备注/Memo:
收稿日期:2016-11-19;改回日期:。
基金项目:国家自然科学基金项目(61472095,61502116);黑龙江省教育厅智能教育与信息工程重点实验室开放基金项目.
作者简介:滕旭阳,男,1987年生,博士研究生,主要研究方向为机器学习、智能优化算法;董红斌,男,1963年生,教授,博士生导师,主要研究方向为多智能体系统、机器学习;孙静,女,1993年生,硕士研究生,主要研究方向为机器学习、数据挖掘。
通讯作者:孙静.E-mail:sunjing@hrbeu.edu.cn.
更新日期/Last Update: 1900-01-01