The biggest mistake that data scientists make is the classic case of correlation vs. causation. Just because two trends are linked, it doesn't mean that one causes another. An example of this that I'm aware of is a tech company that developed an application to sort resumes. The application recognized from the data that most engineers that ended up being hired were male. Of course, this is a correlation, because most applicants are male. But the software "taught" itself that it was causation—i.e. being male caused them to be good candidates. So the program started automatically rejecting female applicants. When the company found out, they had to shut down the program.
© 2019 Praveen Puri
Praveen Puri is the Strategic Simplicity® expert who has delivered over $400 million in value. He helps clients "weaponize" simplicity and bridge the gap between strategy and execution. Visit PuriConsulting.com