10 years ago
It is a Mistake to…. Answer Every Inquiry

 (Part 9 (of 11) of the Top 10 Data Mining Mistakes, drawn from the Handbook of Statistical Analysis and Data Mining Applications) I’m tempted to start with a kind of query that experience teaches some of us not to answer, like “Does this data make me look fat?” But that actually misleads about the point I’m trying to make. Data Scientists (and their models) should answer all queries as truthfully as the evidence allows, regardless of how happy or unhappy that makes the questioner. What I am arguing here is we shouldn’t answer when our opinion is unqualified;