Handwritten Digit Prediction using Machine Learning Algorithms

Abstract

People's expectations for machines have never been higher; Everything from classifying objects in images to adding sound to silent movies can be done with the help of deep learning (DL) and machine learning (ML) algorithms. Similar to that, handwriting prediction is a significant field of study and development with numerous accomplishments. A computer's capacity to identify and decode handwriting from images, documents, touch screens, and other media is called handwriting recognition (HWR), often referred to as handwriting text recognition (HTR). Obviously, with the aid of the MNIST dataset, we code using Support Vector Machine (SVM), Multilayer Perceptron (MLP), and Convolutional Neural Network (CNN) models in this article. To determine the best-known number, our primary objective is to compare the accuracy and execution times of the aforementioned models. Keywords: deep learning, machine learning, coding, MNIST dataset, support vector machine (SVM), multilayer perceptron (MLP) and convolutional neural network (CNN). Training, machine learning, numerical prediction, MNIST dataset, Support Vector Machines (SVMs), Multilayer Perceptron (MLPs) and Convolutional Neural Networks (CNN).