Analysis of Resampling Techniques on Classification Models-a Case Study

Supervised learning is one of the machine learning techniques which is broadly classified into Regression and Classification Algorithms.This paper focuses on two models Logistic Regression and Random Forest Classifier.In order to understand the effectiveness of the models, model validation, performance estimation, and addressing imbalanced datasets resampling methods can be used in conjunction with the Logistic Regression and Random Forest Classifier.Choosing and comparing the performances of various resampling methods using different data sets on both the classification models.The need for analysing the different resampling method arises as in the realworld scenarios, due to the advancement of the technology the number of variables in regression model is often too large and imbalanced.The study involves the analysis on both balance class distribution and training sets that are unbalanced.