Person and Gender Identification from Handwriting Using Machine Learning

Abstract

Identification of an individual and gender from handwritten documents presents an intriguing research problem for researchers, as there has been relatively little research in this area. This research aims to examine a machine learning classification algorithm for recognizing the attributes and handwritten characters of various authors. The study proposes a scheme for user authentication that is based on data from pen tablets and handwriting. This research employed two techniques to ascertain the author's identity and to recognize handwritten characters. According to the study's analysis of experimental findings, the accuracy levels for SVM, LR, LDA, and RF are 87% 85%, 73%, and 77%, respectively. Both SVM and LR have an accuracy level greater than 80%.