Improved Arabic Characters Recognition by Combining Multiple Machine Learning Classifiers

Alabbas, Maytham; Khudeyer, Raidah; Jaf, Sardar; Dong, Minghui; Tseng, Yuen-Hsien; Lu, Yanfeng; Yu, Liang-Chih; Lee, Lung-Hao; Wu, Chung-Hsien; Li, Haizhou

Abstract

In this paper, we investigate a range of strategies for combining multiple machine learning techniques for recognizing Arabic characters, where we are faced with imperfect and dimensionally variable input characters. Experimental results show that combined confidence-based backoff strategies can produce more accurate results than each technique produces by itself and even the ones exhibited by the majority voting combination.

Citation

Alabbas, M., Khudeyer, R., Jaf, S., Dong, M., Tseng, Y.-H., Lu, Y., Yu, L.-C., Lee, L.-H., Wu, C.-H., & Li, H. (2016, November). Improved Arabic Characters Recognition by Combining Multiple Machine Learning Classifiers. Presented at The 20th International Conference on Asian Language Processing., Tainan, Taiwan

• Presentation Conference Type: Conference Paper (published)
• Conference Name: The 20th International Conference on Asian Language Processing.
• Start Date: Nov 21, 2016
• End Date: Nov 23, 2016
• Acceptance Date: Aug 28, 2016
• Online Publication Date: Mar 13, 2017
• Publication Date: Mar 13, 2017
• Deposit Date: Oct 21, 2016
• Publicly Available Date: Oct 24, 2016
• Publisher: Institute of Electrical and Electronics Engineers
• Pages: 262-265
• Book Title: Proceedings of the 2016 International Conference on Asian Language Processing (IALP), 21-23 November 2016, Tainan, Taiwan.
• DOI: https://doi.org/10.1109/ialp.2016.7875982
• Public URL: https://durham-repository.worktribe.com/output/1149517