In this section, we will show you how to use tree-based methods to predict who will click on a mobile advertisement given a set of conditions, such as region, where the advert is shown, time of day, location of the banner, and the application delivering the advertisement.
The dataset we will use throughout the rest of the chapter is obtained from Shioji, Enno, 2017, Adform click prediction dataset, https://doi.org/10.7910/DVN/TADBY7, Harvard Dataverse, V2.
The main task is to build a classifier capable of predicting whether a user will click on an advertisement given certain conditions. Having such a model is very useful for ad-tech platforms that select which ads to show to users and when. These platforms can use these models to only show ads to users who are likely to click on the ad being delivered.
The dataset is large enough (5 GB) to...
