Interpreting Random Forests

me@ryanorban.com (Ryan Orban) — Tue, 23 Dec 2014 00:00:00 +0000

Summary

Random forests are accurate but famously opaque — you get a prediction but not a clear explanation of why. The datadive.net blog post addresses this by walking through the techniques available at the time (2014) for extracting insight from trained random forest models. This was a live research concern: random forests were the workhorse ensemble method of early-2010s machine learning, and the interpretability gap versus simpler models like logistic regression was a genuine obstacle to deployment in regulated industries.

Random-Forests on Ryan Orban

Interpreting Random Forests

Summary