DSCI100 Project Group50

Group Member: Zihang Cheng (Allen), NingDing (Dina), Ruohan Sun (Sabrina), Jiahao Li (Peter)

Exploring Optimal Models for Recognizing Handwritten English Characters: A Data Science Approach

Introduction

$\qquad$ In the modern day, recognizing and interpreting handwritten characters have played a crucial role in our life. Accurately recognizing characters leads to a wide range of application areas, including automatic recognizing english characters, which improved artificial intelligence (Dhande & Kharat, 2017). This research aims to develop optimal models for the recognition and interpretation of handwritten English characters in a given picture. To achieve this, we employ classification and linguistic regression in the $R$ programming language and utilize visualization to create various plots to present our research findings.

$\qquad$ The major goal of our research is to create an effective approach for recognizing and interpreting handwritten English characters (upper case A-Z). The benefit of creating a reliable computer handwriting recognition for the $26$ letters of the alphabet can greatly reduce the need for manual transcribing, which leads to significant time savings and reduces labor costs in various industries that heavily rely on handwritten documents. Meanwhile, handwriting recognition can play a vital role in medical caring. For example, forming communication with aphasiac.

$\qquad$ The Research of handwriting recognition in computers can prompt the advancement of artificial intelligence and technology. Techniques like deep learning, convolutional neural networks, or recurrent neural networks, which are commonly used for handwriting recognition, could benefit from the findings and methods presented in the project. This could lead to further innovation and advancement in the broader field of pattern recognition and computer vision.