Citation
Hossen, Md Ismail and Hossen, Md. Jakir and Hossain, Md Nazmul (2022) An Approach to Recognize Handwritten Digits Using Machine Learning Classifiers. Lecture Notes in Electrical Engineering, 835. pp. 449-458. ISSN 1876-1100 Full text not available from this repository.Abstract
Handwritten digit recognition is one of the most important issues in the area of pattern recognition researches. There are many uses of handwritten digit recognition such as Bank check processing, sorting postal mail form, phone number data entry are common applications of automatic digit recognition. The sentiment of the problem deceits within the capability to develop an efficient algorithm that can recognize handwritten digits. Usually, these digits are normally found from scanning documents with digital devices. Typically storing handwritten digits such as phone number, bank account number, postal numbers and so is extremely troublesome with human intervention. An efficient handwritten digit recognition can eradicate this hazard. To eliminate the difficulties of recognizing handwritten digits, this paper proposes an approach using machine learning algorithms. The objective of this research is to present a reliable and effective approach to recognize handwritten digits. Several supervised machine learning classifiers were employed for the recognition and their accuracy are compared and discussed. The highest 97.07% accuracy is found by the Random Forest classifier.
Item Type: | Article |
---|---|
Uncontrolled Keywords: | Handwritten digit recognition, Machine learning, Text recognition |
Subjects: | Q Science > Q Science (General) > Q300-390 Cybernetics |
Divisions: | Faculty of Engineering and Technology (FET) |
Depositing User: | Ms Nurul Iqtiani Ahmad |
Date Deposited: | 09 Aug 2022 03:16 |
Last Modified: | 09 Aug 2022 03:16 |
URII: | http://shdl.mmu.edu.my/id/eprint/10182 |
Downloads
Downloads per month over past year
Edit (login required) |