Recognition of Handwritten Numbers Using Machine Learning and Deep Learning

EasyChair Preprint 9553

Abstract

People are striving to use computers to complete the majority of their work as technology is improving daily. The process may involve classifying objects in pictures or giving silent movies sound. Deep learning and machine learning algorithms are able to do any of the above tasks. Similarly, recognizing handwritten text is a significant area of research and development with several potential applications. Handwriting recognition is the capacity of a system to identify & decode comprehensible handwritten input from various types of sources. Evidently, utilizing the MNIST datasets and the K Nearest Neighbor Algorithm(KNN), Support Vector Machines(SVM) Algorithm, Random Forest Classifier(RFC) Algorithm, and Convolution Neural Network models(CNN), we have provided an outline of handwritten digit recognition in this research. The accuracy of the models mentioned above, as well as how well they were implemented, will be compared in order to obtain the most reliable results.

Keyphrases: Convolutional Neural Network, Handwritten Digit Recognition, K-Nearest Neighbour, MNIST dataset, Random Forest Classifier, Support Vector Machine, deep learning, machine learning

@booklet{EasyChair:9553,
  author    = {M Surekha and Riya Goyal and Ritika Sahu and Sumeet Singh},
  title     = {Recognition of Handwritten Numbers Using Machine Learning and Deep Learning},
  howpublished = {EasyChair Preprint 9553},
  year      = {EasyChair, 2023}}