A novel AdaBoost algorithm based on rough set

被引：0

作者：

Cheng, Shunkuan ^{[1
]}

Xu, Su ^{[1
]}

Tu, Wenhua ^{[1
]}

机构：

[1] Institute of Information Engineering, Nanchang University, Nanchang

来源：

Journal of Information and Computational Science | 2015年 / 12卷 / 09期

关键词：

AdaBoost; Classifier; Rough Set; Weight Distribution;

D O I：

10.12733/jics20106054

中图分类号：

学科分类号：

摘要：

AdaBoost has strong learning capability by training weak classifier iteratively, of which the key lies in updating the weights of samples. However, when the weights of samples misclassified are greater than a certain value, the generated basic classifier will fluctuate leading to a large error in the results. To avoid this, a new method combining AdaBoost with Rough set is proposed. The method adopts an advanced optimization factor and is verified to be feasible theoretically. Firstly, it applies Rough set to split the misclassified samples into two kinds: one is very likely to have been classified correctly in the current iteration and the other not, then readjusts the weight distribution to improve the latter's probability of being classified correctly. The simulation results show that: compared with traditional AdaBoost, the proposed method can not only converge more quickly but also reduce the error globally. ©, 2015, Journal of Information and Computational Science. All right reserved.

引用

页码：3485 / 3494

页数：9

共 14 条

[1]

Freund Y., Schapire R.E., A decision-theoretic generalization of on-line learning and an application to boosting, Journal of Computer and System Scienses, 55, 1, pp. 119-139, (1997)

[2]

Miao D., Zheng H., Gu G., AdaBoost face detection algorithm based on optimized weighting parameter, Computer Engineering and Applications, 50, 19, pp. 173-177, (2014)

[3]

Fang Y., Lu D., Ge J., Method for license plate detection based on improved AdaBoost algorithm, Application Research of Computers, 26, 12, pp. 4827-4829, (2009)

[4]

Liao H., Zhou D., Review of AdaBoost and its improvement, Computer Systems & Applications, 21, 5, pp. 240-244, (2012)

[5]

Fu Z., Effectiveness analysis of AdaBoost, Journal of Computer Research and Development, 45, 10, pp. 1747-1755, (2008)

[6]

Li H., Illtrusion detection method based on rough set and adaptive boost, Journal of Computer Applications, 29, 1, pp. 181-183, (2009)

[7]

Wang G., Rough Set Theory and Knowledge Acquisition, (2001)

[8]

Sun S., Yao L., Wu Q., Et al., Research of generalized variable precision rough set model and its application, Computer Engineering and Applications, 45, 7, pp. 10-13, (2009)

[9]

Yang C., Deng T., An approach to attribute reduction combining attribute selection and deletion, CAAI Transactions on Intelligent Systems, 8, 2, pp. 183-186, (2013)

[10]

Wang G., Yu H., Yang D., Decision table reduction based on conditional information entropy, Chinese Journal of Computers, 25, 7, pp. 759-766, (2002)

← 1 2 →