From the course: Predictive Analytics Essential Training: Data Mining

Beginning with a solid first step: Problem definition

From the course: Predictive Analytics Essential Training: Data Mining

Start my 1-month free trial Buy for my team

Beginning with a solid first step: Problem definition

“

- When most analysts try to build predictive models, they too often take a research request uncritically and then accept data uncritically, and then run at high speed to their modeling tool of choice and start modeling. Poor problem definition is almost certainly the biggest reason that projects fail. If you are very lucky, the project will merely run late, and you'll eventually get where you need to go, but there is a very real possibility that it will fail completely, and the likelihood that the project will result in a valuable deployed solution will be almost nil. I'm not trying to be overly pessimistic, I've just seen it too many times. Let's learned about the four essential elements of problem definition so that you won't make these mistakes.

Contents