STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Operating Systems Development & Trends

Identification of Handwritten Digits using Machine Learning

  1. Devansh Gera
  2. Yash Popli
  3. Gungun Singhal
  4. Komal Mittal
  5. Reena Sharma

Abstract

Handwritten Digit Recognition is one of the practical issues in sample recognition applications. The task for handwritten recognition has been difficult due to various variations in written styles. The capacity to create an effective algorithm that can detect handwritten numbers given by users via a scanner, tablet, and other digital devices is at the core of the issue. Artificial intelligence is used in machine learning, which automatically corrects errors based on experience. In this paper, we present a proposal to off-line Handwritten Digit Recognition through machine learning techniques. This article compares classifiers like KNN, PSVM, NN, and convolution neural network on the basis of performance, accuracy, time, sensitivity, positive productivity, and specificity while using different parameters with the classifiers. The handwritten digits (0 to 9) from the well-known Modified National Institute of Standard and Technology (MNIST) dataset are recognized.

Keywords

Support