Bid Optimization for Real-Time Bidding (RTB) Online Advertising

Please enable javascript in your browser.

Gheshlaghpour, Fatemeh | 2018

1049 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 51194 (05)
University: Sharif University of Technology
Department: Electrical Engineering
Advisor(s): Nobakhti, Amin; Babazadeh, Maryam
Abstract:
Real-Time Bidding (RTB) has revolutionized the online display advertising market in recent years. In RTB market the advertisers are able to use impression-level data, such as user cookies and context information, to bid for each ad impression in the RTB auctions.Therefore, it is important for an advertiser to design an effective bidding strategy. The proposed bidding strategy should consider the target audience of the campaign and the information of each specific ad impression.This thesis is focused on the design of optimized bidding strategies for RTB auctions.By the optimal strategy, we mean the one which maximizes the campaign’s Key Performance Indicator (KPI) in the presence of budget constraint, from the perspective of an advertiser. In this work, the problem is mathematically formulated as a standard optimization, considering the total clicks of the campaign as KPI and a series of effective bidding strategies are proposed. A logistic estimator is also adopted to predict the user response.Comparison with the widely used bidding strategies in product, confirms that the proposed bidding methods overally outperform the existing methods on total clicks
Keywords:
Optimal Bidding Behavior Model ; Logistic Regression (LR)Analysis ; Display Advertisting ; Real-Time Bidding (RTB)