Comparing Error Rates of MNIST Datasets Using Various Machine Learning Models

EasyChair Preprint 13152

Abstract

This study investigates the performance of different machine learning models on the MNIST dataset, a widely used benchmark dataset for handwritten digit recognition. The aim is to compare the error rates of various models to determine their effectiveness in accurately classifying digits. Four machine learning models, namely Logistic Regression, Support Vector Machine, Random Forest, and Convolutional Neural Network, are implemented and evaluated. The models are trained and tested using the MNIST dataset, and their error rates are compared. The results show that the Convolutional Neural Network outperforms the other models, achieving the lowest error rate. This study provides valuable insights into the performance of different machine learning models on the MNIST dataset, which can be useful for researchers and practitioners working in the field of pattern recognition and machine learning.

Keyphrases: AI, deep learning, machine learning

@booklet{EasyChair:13152,
  author    = {Favour Olaoye},
  title     = {Comparing Error Rates of MNIST Datasets Using Various Machine Learning Models},
  howpublished = {EasyChair Preprint 13152},
  year      = {EasyChair, 2024}}