Some people think machine learning models are black boxes, useful for making predictions but otherwise unintelligible; but the best data scientists know techniques to extract real-world insights from any model.
Why are these important? The five most important applications of model insights are Debugging, informing feature engineering, directing future data collection, informing human decision-making, Building Trust
Debugging is one of the most valuable skills in data science. Understanding the patterns, definitely helps one identify when those are against our knowledge of the real world, and this is basically the first step in tracking down bugs.
Informing Feature Engineering
Feature engineering usually involves repeatedly creating new features using modifications to our raw data or features we had previously created.
Directing Future Data Collection
Model-based insights give you a good understanding of the value of features you currently have, which will help you reason about what new values may be most helpful.
Informing Human Decision-Making
Many important decisions are made by humans. For these decisions, insights can be more valuable than predictions.
Building Trust
Many people won’t trust one’s model for important decisions without verifying some basic facts. This is a smart precaution given the common presence of data errors.
Originally published at https://www.jtpcindia.com.
