Comparison of ML classifiers for Image Data

EasyChair Preprint 3815

Abstract

In this paper, we have compared performance of various machine learning classifiers such as Multinomial Logistic Regression, Support Vector Machine, Multi Layer Perceptron, Random Forests, Naive Bayes, K Nearest Neighbors, ADA Boost and Convolutional Neural Networks on 2 popular image data sets CIFAR-10 and MNIST. Then we tried to find out reason of performance variance. We also considered the significant of feature extraction and feature selection. Convolutional Neural Networks has been winner over all other ML classifiers. We found out that CNN performs feature extraction and selection automatically which no other classifier is able to do.

Keyphrases: Artificial Intelligence, computer vision, feature extraction, feature selection, image classification, machine learning

@booklet{EasyChair:3815,
  author    = {Sonika Dahiya and Rohit Tyagi and Nishchal Gaba},
  title     = {Comparison of ML classifiers for Image Data},
  howpublished = {EasyChair Preprint 3815},
  year      = {EasyChair, 2020}}