This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
Machine learning is both exciting and daunting. We’ve all heard stories of some of the problems that have come up as ML has flourished, and yes, there can be risks. However, there’s also immense possibility for using these tools in productive, positive and exciting ways.
At the Virtual Global Inspire event this year, I presented three sessions — Machine Learning 101, 102 and 103 — with the goal of helping anyone become more comfortable and familiar with the tools of machine learning. I’m really happy to share these with you here so they can be a resource as you explore the possibilities of ML.
In Machine Learning 101, we cover the basics: defining machine learning, examining the kinds of questions that machine learning can (and can’t) help address, reviewing the kinds of data we need, and looking in a bit more detail at a few potential use cases.
In Machine Learning 102, we cover how a machine learning workflow takes shape, including the critical steps of feature engineering and feature selection. We also review how models work in general terms — and how we can measure how well they’re working.
Machine Learning 103 wraps up this series with a discussion of the potential “potholes and landmines” in the machine learning process, including keys to successful data prep, critical things to consider in feature selection, and avoidance of target leakage.
Thanks for checking out this series of videos! I hope this has been a useful introduction (or review) of some of the key principles and processes of machine learning.
I go into a lot more detail about these and many more topics in the Education Mode text that’s built into our new ML Automation product, Alteryx Machine Learning (AML). We have specifically developed AML to be approachable for beginners to ML. It not only guides you through the process of building models, but it also performs many types of analysis on your data and models in order to help ensure that you avoid the types of potential issues that I outlined in the ML 103 talk, so you get reliable results.
Alteryx Machine Learning is currently in Limited Availability, and will be Generally Available in a few months. I encourage you to think through some use cases that would benefit you, and to give it a try as soon as possible!
Watch this space for more resources yet to come.
What questions do you have about getting started with machine learning? What are your concerns, and what gets you excited about its potential? Let us know with a comment, and be sure to subscribe to the blog to get future articles.