Here are four reasons why the result of your analytics modeling might be correct (according to some accuracy metric), but it might not be the right answer:
Your model may be underfit.
Your model may be overfit.
Your model may be biased.
- Your model may be suffering from the false positive paradox.
In data science, we are trained to keep searching even after we find a model that appears to be accurate. Data Scientists should continue searching for a better solution, for at least the four reasons listed above. Please note that I am not advocating “paralysis of analysis”, where never-ending searches for new and better solutions are just an excuse (or a behavior pattern) that prevents one from making a final decision. Good leaders know when an answer is “good enough”. We discussed this in a previous article: “Machine Unlearning – The Value of Imperfect Models”…
(For more discussion of the four cases listed above, continue reading here … https://www.mapr.com/blog/4-reasons-look-further-accurate-answer-your-analytics-question)
Follow Kirk Borne on Twitter @KirkDBorne