• Anjesh Kumar
• Sukrant Kathuria
• Mrs. Shikha Gupta
• Dr. Atul Mishra

Online Advertising is the main source of income in Digital Market as the internet is used widely today. Although, there are so many ways for advertising but because of the popularity of the internet, digital advertising became very popular and effective to earn revenue. But it also faces many problems. Bidding optimization is one of the most crucial problems in online advertising. Keyword-level bidding techniques are used in Sponsored search (SS) auction processes, because of the randomness of user’s query behavior and platform nature. But the display advertising (DA) has taken benefits of real-time bidding (RTB) process to improve the performance for advertisers. In our work, we consider the RTB problem in sponsored search (SS) auction, named as SS-RTB. SS-RTB has a more complicated dynamic behavior, due to stochasticness of user’s query behavior and more complicated bidding prices policies based on multiple keywords of an advertisement. Most previous techniques for Digital Advertisement can’t be applied. Here analyze a reinforcement learning (RL) method for handling the complex dynamic behavior of bidding. Here, the state transitions fluctuate between two days. By observation of sequences, formulate a robust MDP model according to time i.e., hour-aggregation level of the auction’s data and developed a model for SS-RTB. By not generating bid prices directly, Robust MDP model is used for impressions of hours basis and performed real-time bidding.

Real-time bidding,computational advertising, Reinforcement Learning (RL), Sponsored Search (SS), Markov Decision Process (MDP).

"A Survey on Bidding Optimization Techniques in Online Advertising", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 3, page no.655-663, March-2019, Available :http://www.jetir.org/papers/JETIR1903789.pdf

"A Survey on Bidding Optimization Techniques in Online Advertising", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 3, page no. pp655-663, March-2019, Available at : http://www.jetir.org/papers/JETIR1903789.pdf