Feature Selection Using a Reinforcement-Behaved Brain Storm Optimization

In this era of data explosion, feature selection has received sustained attention to remove the large amounts of meaningless data and improve the classification ac-curacy rate. In this paper, a feature selection method based on reinforcement-behaved strategy is proposed, which identifies the most important features by embedding the Brain Storm Optimization (BSO) algorithm into the classifier. The ideas of the BSO are mapped to feature subsets, and the importance of the feature is evaluated through some indicators, i.e. the validity of the feature migration. In the migration of each feature, the feature is updated to a new feature in the same position between the two generations. The feedback of each action is used as the basis for the ordering of feature importance. An updating strategy is presented to modify the actions based on the current state to improve the feature set. The effectiveness of the proposed algorithm has been demonstrated on six different binary classification datasets (e.g., biometrics, geography, etc.) in comparison to several embedded methods. The results show that our proposed method is superior in high performance, stability and low computing costs.

[1]  Zehang Sun,et al.  Object detection using feature subset selection , 2004, Pattern Recognit..

[2]  Nasser M. Nasrabadi,et al.  Pattern Recognition and Machine Learning , 2006, Technometrics.

[3]  Isabelle Guyon,et al.  An Introduction to Variable and Feature Selection , 2003, J. Mach. Learn. Res..

[4]  Ben Niu,et al.  A discrete bacterial algorithm for feature selection in classification of microarray gene expression cancer data , 2017, Knowl. Based Syst..

[5]  Yuhui Shi,et al.  Optimal Satellite Formation Reconfiguration Based on Closed-Loop Brain Storm Optimization , 2013, IEEE Computational Intelligence Magazine.

[6]  Mengjie Zhang,et al.  Variable-Length Particle Swarm Optimization for Feature Selection on High-Dimensional Classification , 2019, IEEE Transactions on Evolutionary Computation.

[7]  Yuhui Shi,et al.  Brain Storm Optimization Algorithm , 2011, ICSI.

[8]  Gang Wang,et al.  A novel bacterial foraging optimization algorithm for feature selection , 2017, Expert Syst. Appl..

[9]  Jon Atli Benediktsson,et al.  Feature Selection Based on Hybridization of Genetic Algorithm and Particle Swarm Optimization , 2015, IEEE Geoscience and Remote Sensing Letters.

[10]  Akshya Swain,et al.  A two-dimensional (2-D) learning framework for Particle Swarm based feature selection , 2018, Pattern Recognit..

[11]  Parham Moradi,et al.  Integration of graph clustering with ant colony optimization for feature selection , 2015, Knowl. Based Syst..

[12]  Yaochu Jin,et al.  Feature selection for high-dimensional classification using a competitive swarm optimizer , 2018, Soft Comput..

[13]  Hossein Nezamabadi-pour,et al.  An advanced ACO algorithm for feature subset selection , 2015, Neurocomputing.

[14]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[15]  Huan Liu,et al.  Feature Selection for Classification , 1997, Intell. Data Anal..

[16]  Sunanda Das,et al.  Ensemble feature selection using bi-objective genetic algorithm , 2017, Knowl. Based Syst..

[17]  Zhi-hui Zhan,et al.  A modified brain storm optimization , 2012, 2012 IEEE Congress on Evolutionary Computation.

[18]  Guy Lapalme,et al.  A systematic analysis of performance measures for classification tasks , 2009, Inf. Process. Manag..

[19]  Yudong Zhang,et al.  Binary PSO with mutation operator for feature selection using decision tree applied to spam detection , 2014, Knowl. Based Syst..

[20]  Robert Tibshirani,et al.  The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd Edition , 2001, Springer Series in Statistics.