Object Classification Method Using Dynamic Random Forests and Genetic Optimization

김재협; 김헌기; 장경현; 이종민; Young Shik Moon (문영식)

Object Classification Method Using Dynamic Random Forests and Genetic Optimization

Journal of The Korea Society of Computer and Information
Abbr : JKSCI
2016, 21(5), pp.79-89
Publisher : The Korean Society Of Computer And Information
Research Area : Engineering > Computer Science

김재협 ¹, 김헌기 ¹, 장경현 ¹, 이종민 ², Young Shik Moon ²

¹한화탈레스
²한양대학교

Accredited

ABSTRACT

In this paper, we proposed the object classification method using genetic and dynamic random forest consisting of optimal combination of unit tree. The random forest can ensure good generalization performance in combination of large amount of trees by assigning the randomization to the training samples and feature selection, etc. allocated to the decision tree as an ensemble classification model which combines with the unit decision tree based on the bagging. However, the random forest is composed of unit trees randomly, so it can show the excellent classification performance only when the sufficient amounts of trees are combined. There is no quantitative measurement method for the number of trees, and there is no choice but to repeat random tree structure continuously. The proposed algorithm is composed of random forest with a combination of optimal tree while maintaining the generalization performance of random forest. To achieve this, the problem of improving the classification performance was assigned to the optimization problem which found the optimal tree combination. For this end, the genetic algorithm methodology was applied. As a result of experiment, we had found out that the proposed algorithm could improve about 3~5% of classification performance in specific cases like common database and self infrared database compare with the existing random forest. In addition, we had shown that the optimal tree combination was decided at 55~60% level from the maximum trees.

KEYWORDS

Object Classification, Random Forest, Genetic Algorithm, Classifier Ensemble

Citation status

* References for papers published after 2023 are currently being built.

[book] D. E. Rumelhart / 1986 / Parallel distributed processing: explorations in the microstructure of cognition / MIT Press

[journal] C. J. C. Burges / 1998 / A tutorial on support vector machines for pattern recognition / Journal of Data Mining and Knowledge Discovery 2 : 121~167

[journal] 염해영 / 2005 / Classification of Gene Data Using Membership Function and Neural Network / 전자공학회논문지 - CI / 대한전자공학회 42(4) : 251~260

[journal] 강선경 / 2008 / Enhancement of the Correctness of Marker Detection and Marker Recognition based on Artificial Neural Network / 한국컴퓨터정보학회논문지 / 한국컴퓨터정보학회 13(1) : 89~98

[journal] 김광성 / 2010 / Support Vector Machine Algorithm for Imbalanced Data Learning / 한국컴퓨터정보학회논문지 / 한국컴퓨터정보학회 15(7) : 11~17

[journal] 김재협 / 2014 / Gunnery Classification Method Using Profile Feature Extraction in Infrared Images / 한국컴퓨터정보학회논문지 / 한국컴퓨터정보학회 19(10) : 43~53

[journal] Robi Polikar / 2006 / Ensemble based systems in decision making / IEEE Circuit and Systems 6(3) : 21~45

[journal] P. Viola / 2004 / Robust real-time face detection / Int. Journal of Computer Vision 57(2) : 137~154

[journal] L. Breiman / 2001 / Random forest / Machine Learning 45 : 5~32

[journal] 김재협 / 2010 / Multi-target Classification Method Based on Adaboost and Radial Basis Function / 전자공학회논문지 - CI / 대한전자공학회 47(3) : 22~28

[journal] 정경중 / 2014 / Facial Expression Recognitionusing Face Alignment and AdaBoost / 전자공학회논문지 / 대한전자공학회 51(11) : 193~202

[other] L. Breiman / 1993 / Chapman and Hall

[confproc] R. Banfield / 2004 / A comparison of ensemble creation technique / Proc. of Multiple Classifier Systems 1 : 223~232

[confproc] S. Bernard / 2009 / Influence of hyperparameters on random forest accuracy / Proc. of Workshop on Multiple Classifier Systems 1 : 171~180

[confproc] S. Bernard / 2009 / On the selection of decision trees in random forest / Proc. of Joint Conf. on Neural Networks : 302~307

[confproc] E. Tripoli / 2010 / Dynamic construction of random forests: Evaluation using biomedical engineering problems / Proc. of IEEE Int. Conference on Information Technology and Application in Biomedicine 1 : 3~5

[journal] S. Bernard / 2012 / Dynamic random forests / Journal of Pattern Recognition Letters 33(12) : 1580~1586

[confproc] F. Roli / 2001 / Methods for designing multiple classifier systems / Proc. of 2nd International Workshop MCS2001 : 78~87

[journal] I. S. Oh / 2004 / Hybrid genetic algorithms for feature selection / IEEE Trans. on Pattern Analysis and Machine Intelligence 26(11) : 1424~1437

[journal] 허기수 / 2009 / Genetic Algorithm for Node Pruning of Neural Networks / 전자공학회논문지 - CI / 대한전자공학회 46(2) : 65~74

[web] / http://archive.ics.uci.edu/ml/ / http://archive.ics.uci.edu/ml/

[journal] 김철문 / 2013 / An Efficient Pedestrian Recognition Method based on PCA Reconstruction and HOG Feature Descriptor / 전자공학회논문지 / 대한전자공학회 50 : 162~170

[confproc] R. Maclin / 1998 / Boosting Classifiers Regionally / Proc. of the 15th National Conference on Artificial Intelligence 1 : 700~705

[journal] Venkatadri M. / 2010 / A multiobjective genetic algorithm for feature selection in data mining / Journal of Computer Science and Information Technology 1(5) : 443~448

[journal] T. G. Dietterich / 2010 / An experimental comparison of three methods for constructing ensembles of decision trees : bagging, boosting, and randomization / Journal of Machine Learning 40(2) : 139~157

KJCKorea
Journal Central

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65