A practical approach to feature selection
ML92 Proceedings of the ninth international workshop on Machine learning
Computational Statistics & Data Analysis - Nonlinear methods and data mining
Predicting clicks: estimating the click-through rate for new ads
Proceedings of the 16th international conference on World Wide Web
Estimating rates of rare events at multiple resolutions
Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining
Adaptive bidding for display advertising
Proceedings of the 18th international conference on World wide web
A search-based method for forecasting ad impression in contextual advertising
Proceedings of the 18th international conference on World wide web
Handling forecast errors while bidding for display advertising
Proceedings of the 21st international conference on World Wide Web
Dynamic ad layout revenue optimization for display advertising
Proceedings of the Sixth International Workshop on Data Mining for Online Advertising and Internet Economy
Ad impression forecasting for sponsored search
Proceedings of the 22nd international conference on World Wide Web
Predicting advertiser bidding behaviors in sponsored search by rationality modeling
Proceedings of the 22nd international conference on World Wide Web
Forecasting user visits for online display advertising
Information Retrieval
Hi-index | 0.00 |
Display advertising has been a significant source of revenue for publishers and ad networks in online advertising ecosystem. One important business model in online display advertising is Ad Exchange marketplace, also called non-guaranteed delivery (NGD), in which advertisers buy targeted page views and audiences on a spot market through real-time auction. In this paper, we describe a bid landscape forecasting system in NGD marketplace for any advertiser campaign specified by a variety of targeting attributes. In the system, the impressions that satisfy the campaign targeting attributes are partitioned into multiple mutually exclusive samples. Each sample is one unique combination of quantified attribute values. We develop a divide-and-conquer approach that breaks down the campaign-level forecasting problem. First, utilizing a novel star-tree data structure, we forecast the bid for each sample using non-linear regression by gradient boosting decision trees. Then we employ a mixture-of-log-normal model to generate campaign-level bid distribution based on the sample-level forecasted distributions. The experiment results of a system developed with our approach show that it can accurately forecast the bid distributions for various campaigns running on the world's largest NGD advertising exchange system, outperforming two baseline methods in term of forecasting errors.