You’re looking at your iPhone for a particular picture of your friend, taken a couple of years ago. Within seconds you find the picture you’re looking for.
It can be boiled down to three big hurdles: the wrong data in the wrong infrastructure at the wrong time.
New feature engineering — processing data necessary to choose and train models — requires going back to the raw data for different aggregations.
The quality of the models, and their outcomes, increases with the volume of event data ingested.