There is an even faster increase of ad blocker usage, which influence badly on publishers’ and advertisers’ business. Thus, more and more companies initialize their counter-ad blocking strategies, in which customers choose to either disable their ad blockers or leave without seeing the content. For example, Forbes and LA times. Ad blocker usage causes $40 billion revenue lost for publishers and it puts a severe threat to the existing ad-supported Free Web Content.
What it does
Our goal is to predict the whitelist prediction (whitelist means to turn off ad blockers) given a user intends to view one new article. These predictions can help publishers better understand ad blocker usage, enhance customer relationship, and further provide suggestions to improve users’ whitelist propensity, so as to keep the business and maintain the Free Web Content!
How I built it
I built it with python using sklearn library.
Challenges I ran into
Web data has a lot of noise and outliers. Most features are categorial variables.
Accomplishments that I'm proud of
Proposed the random forest model to handle categorial variables. The studied problem is very important and significant.
What I learned
How to handle categorial variables? How to tune the parameters in random forests?
What's next for AdBlocker_Whitelist_Prediction_in_Online_News_Reading
Data visualization for the result. Deploy the model into AWS and provide API to the public.