Comparative study of select non parametric and ensemble machine learning classification techniques

Authors:

Abstract
Supervised classification is a machine learning approach where data is classified among various discrete classes which could be binary or multi-class. Decision Tree, random Forest, Rotation Forest, AdaBoost, Extra Tree, XGBoost are some of the ensembles that classify data on trained model under certain criterion. AUROC and in-sample out-sample error is used as performance measure of algorithms on benchmark datasets. Hyperparameter tuning is done to improve performance and avoid overfitting.

This publication has 13 references indexed in Scilit: