Bayesian thinking in work and life, ad attribution models and A/B testing, machine learning@Foursquare - Max Sklar - the data scientist show050
The Data Scientist Show - Daliana Liu
00:00
How Did You Build a Causality Model?
There were definitely a few cases where the causality model was just broken. A few exmples were, think places like a them parks, or tourist ads. We used as features the place that the person was at the time, the day they visited. Like hay, i woke up in salem, new hampshire. So that's a feature. The problem is the ad caused you to travel. Another thing that broke one was a, there was one where, like, the lift was like negative thirt%. And they were asking me, like, what the heck is going on here? Ai was like, no, ad is that bad it turned out to be an ad
Transcript
Play full episode
