CTR Prediction of Advertisements using Decision Trees based Algorithms

Abstract

In this age of digitization, all the businesses have started focusing their attention on getting customers online. In the present scenario to attract huge customer bases, businesses require proper marketing which is incomplete without advertising. To maximize their reach, online advertising came into picture and to optimize their marketing potential, knowing and understanding the CTR(Click Through Rate) of an advertisement is very important. This paper delves into the sector of machine learning, to predict the CTR of an advertisement. It provides a comparative study of four algorithms - Decision Trees, XGB(Extreme Gradient Boosting), Random Forest and LGBM (Light Gradient Boosting Method) - based on their performance to determine which algorithm gives the highest AUC(Area Under the Curve) score, F1 score, accuracy and precision.