Classification Problems in Machine Learning

Abstract

In this report, we will introduce and study the selection and applications of the classifiers in the machine learning. We will use two examples to illustrate how to choose the classifier and how to evaluate the classifier’s generalization performance, so as to choose the best classifier to solve different problems. Case study 1: recognize the handwritten digits, we choose SGD classifier, random forest classifier and KNN classifier to stimulate and predict, in order to evaluate the effect of different classifiers and conclude the advantages and shortages of them. Case study 2: identify the species of IRIS, we will use logistic regression, decision tree and KNN to form an ensemble classifier to identify the species of IRIS so that we can conclude the advantages and shortages for both ensemble and single adjective classifiers.