How Can Machine Learning Find Insights Into My Data? Part 1 of 3

ML automatically finds patterns in data. See how it can help you better understand yours!

Raymond E Peck III
13 min readApr 4, 2022
A Partial Dependence plot showing that employee attrition rises in years 5 and 6
Machine Learning can find insights from your data!

Machine Learning (ML) can be an intimidating term, but it really shouldn’t be. ML is a topic that is very simple to grasp in principle. It’s all about:

pattern recognition!

At its core, ML looks at your data and tries its best to find any patterns in it. It then saves those patterns as a model. Usually models are used to make predictions, but we can also use them to understand these patterns it finds in our data. This can give us insights that we couldn’t get any other way, since ML can identify complex patterns that simple statistical techniques can’t find. Fortunately, it’s pretty straightforward to interpret what it’s telling us with the tools we have available these days. We do have to learn a few things to avoid being mislead, but stick with me here, I’ll cover that in a bit.

A Topic in Three+ Posts

I wrote the original version of this material in one very long post (21 pages / 22 minutes estimated reading time), and it was tightly focused on the Alteryx ML product. I have lengthened the material a bit here on Medium to make it more tool-agnostic…

--

--

Raymond E Peck III

Doing software for ML since 2014: H2O.ai, Quantifind, dotData, and Alteryx. Lots of different kinds of startups since co-founding Pure Software (1995 IPO).