From the course: Predictive Analytics Essential Training: Data Mining

Unlock the full course today

Join today to access over 24,100 courses taught by industry experts.

Seeking surprise results

Seeking surprise results

- [Instructor] One of the things you'll discover is that surprises can be good. I remember on a cell phone loyalty project, we were able to show that a spouse's recent loyalty behavior was a factor. It might sound obvious. A lot of discoveries seem obvious in retrospect. But it wasn't obvious at the time, because it was extremely difficult to bend the data into a position to test it. And no one was particularly focused on it. We only discovered it because we left no stone unturned. So our next element is surprise. A common mistake in data mining is being too frugal with predictors, leaving out this or that variable, because everybody knows that it's not a key driver, or maybe a subject-matter expert told you not to bother because it was not found to be a predictor last time someone tried. Not wise. Even if this is true, it discounts the insight that unanticipated interactions might provide. Variables mix and combine…

Contents