iPinYou Global RTB Bidding Algorithm Competition Dataset

RTB (Real Time Bidding) is one of the most exciting developments in computational advertising in recent years. It drives transparency and efficiency in the display advertising ecosystem and facilitates the healthy growth of the display advertising industry. It enables advertisers to deliver the right message to the right person at the right time, publishers to better monetize their content by leveraging their website audience, and consumers to view relevant information through personalized ads. However, researchers in computational advertising area have been suffering from lack of publicly available datasets. iPinYou organizes a three-season global RTB algorithm competition in 2013. For each season, there is offline stage and online stage. On the offline stage, iPinYou releases a dataset for model training and reserves a dataset for testing. The dataset includes logs of ad biddings, impressions, clicks, and final conversions. After the whole competition ends, iPinYou organizes and releases all these three datasets for public use. These datasets can support experiments of some important research problems such as bid optimization and CTR estimation. To the best of our knowledge, this is the first publicly available dataset on RTB display advertising. In this paper, we give descriptions of these datasets to further boost the interests of computational advertising research community using this dataset.