• A.Lakshmanarao
• G.Vijay Kumar
• K.Chandra Sekhar

Click-through rate (CTR) prediction is one of the important tasks in business applications like online advertising. It is an integral part of online advertising systems. It is used as an input to auctions which find the final ranking of ads. Machine Learning algorithms are very popular to solve interaction problems. Especially Reinforcement Learning algorithms work very well with these kinds of problems. In this paper, we proposed a model based on two algorithms Upper Confidence Bound and Thompson sampling for better click through rate prediction. The dataset is taken from superdatascience which consists of information about 10 ads. If a user click on the ad, reward is 1, otherwise it is 0.We applied reinforcement learning algorithms to this dataset and computed the total rewards for each of the algorithm separately.

Click-Through Rate (CTR), Machine Learning, Reinforcement Learning, Python

"Estimating Click Through Rate Using Reinforcement Learning Algorithms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 3, page no.59-63, March-2019, Available :http://www.jetir.org/papers/JETIRAC06013.pdf

"Estimating Click Through Rate Using Reinforcement Learning Algorithms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 3, page no. pp59-63, March-2019, Available at : http://www.jetir.org/papers/JETIRAC06013.pdf