Offline Handwritten Character Recognition Using Machine Learning

Abstract

In this paper, a comparative analysis of recent techniques for character recognition is done. Our purpose is to identify the impact of machine learning in the domain of character identification. Character recognition has a lot of applications in the fields of banking, healthcare and other fields for searchability, storability, readability, editability, accessibility, etc. to ease up various processes. Traditional machine learning techniques like a neural network, support vector machine, random forest, etc. have been used as classification techniques. Now with the advancement in the field of computer hardware and efficient research in artificial intelligence field have given emergence to deep learning algorithms. Recent articles are using deep learning for character identification. They also depict how various functions improve the performance in the field of pattern recognition over time. The primary purpose of this paper is to encourage young researchers towards this domain and thus learn and work towards achieving novelty in the field of machine learning.