A Novel Model for Recognising Handwritten Devanagari Numerals using Machine Learning

This research study performs a comprehensive comparative analysis aimed at developing effective machine learning models for classifying handwritten Devanagari numerals. The research focuses on evaluating the performance of various models to determine the most accurate classification approach. Initiating with data pre-processing, including feature extraction and normalization, the data is prepared for model training and assessment. A diverse range of machine learning models, from traditional methods like Support Vector Machines (SVM) to advanced techniques such as Random Forests, K-Nearest Neighbors, and Convolutional Neural Networks, are considered for the comparative analysis, ensuring a thorough assessment of classification capabilities. Cross-validation techniques are employed during model training and testing to enhance reliability. Statistical tests are utilized to assess the performance variations among models, enhancing the robustness of the analysis. Visual representations of performance metrics and comparison results offer clear insights. This research study aims to identify the most suitable machine learning model for handwritten Devanagari numeral classification, potentially advancing character recognition systems and linguistic applications.