Ad Prediction using Click Through Rate and Machine Learning with Reinforcement Learning

Abstract

Predicting the click-through rate (CTR) is an essential problem in enterprise systems such as online advertising. It is a crucial factor of advertisements platforms. It is fed into auctions to determine the final ranking of advertising. Machine Learning techniques are often used to tackle challenges involving human-computer interaction. Almost every website on the internet displays advertisements. Companies who want to promote their products use these websites as a method of promotion. The goal is to determine which of the company's several advertisement versions can get the best conversion rate, i.e., the most number of ad clicks. The major issue for firms that rely on ad revenue is ad placement on websites. The placement of the ad has a significant impact on whether or not the ad gets clicked. This kind of challenge lends itself very well to Reinforcement Learning algorithms. In this paper, we applied the machine learning approach for Ad Prediction. We used a dataset from Kaggle and applied two reinforcement learning algorithms Upper Confidence Bound, Thompson Sampling for predicting Ad position based on ad clicks and achieved a good prediction rate. All the implementations are done in python.