doi: 10.4304/jsw.6.4.628-635
Rough Sets and Confidence Attribute Bagging for Chinese Architectural Document Categorization
- School of Information and Control Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China
Abstract
Aiming at the problems of the traditional feature selection methods that threshold filtering loses a lot of effective architectural information and the shortcoming of Bagging algorithm that weaker classifiers of Bagging have the same weights to improve the performance of Chinese architectural document categorization, a new algorithm based on Rough set and Confidence Attribute Bagging is proposed for Chinese architectural document categorization. Rough sets is used to feature selection. First the cores of attributes are found by discernibility matrix and one of the cores is regarded as the start point. Then attributes’ significance and dependency are used as the heuristic information to do feature selection. A Chinese architectural document classifier is designed by Confidence Attribute Bagging algorithm. The voting weights of weaker classifiers are gained by their result and the stronger classifier result is attained by weaker classifiers voting. The algorithm is applied in Attribute Bagging algorithm to design a classifier. The experimental results show that the novel method is not only easy to implement but can effectively reduce the dimensional space, and improve the accuracy of classification.
Keywords
- Rough sets
- feature selection
- Confidenc
How to Cite
Xiang Zhang, Changhua Li, Lili Dong, and Na Ye, "Rough Sets and Confidence Attribute Bagging for Chinese Architectural Document Categorization," Journal of Software, vol. 6, no. 4, pp. 628-635, 2011. https://doi.org/ 10.4304/jsw.6.4.628-635
Copyright & License
Copyright © 2011 by the authors. 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 (CC BY 4.0).