From the course: Machine Learning and AI Foundations: Clustering and Association
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Relating clusters to categories visually
From the course: Machine Learning and AI Foundations: Clustering and Association
Relating clusters to categories visually
- [Instructor] Once you've used some kind of statistical approach to identify which categorical variables might have the strongest relationship to your cluster membership, you can go ahead and display that graphically. The easiest and most effective way to do that is a clustered bar chart. Let me show you. So here I am in Chart Builder, and I'm going to drag up a clustered bar chart. And what I'm going to do is make cluster membership my X, and photo is one of the strongest relationships as it turns out, so I'm going to grab the photo variable and make that my X. And then in the Y, where we've got count here, instead of going with count, I'm going to click on bar one, and say instead that I want percentage. Here's a step, though, that's easy to miss. I'm going to click on Set Parameter, and tell it that I want the percentage for each X-axis category. Click on Continue, and click on OK. Let me describe how this is set up. And then I'll interpret it in the context of photo. What we're…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
-
(Locked)
Relating clusters to categories statistically6m 23s
-
(Locked)
Relating clusters to categories visually2m 45s
-
(Locked)
Running a multiple correspondence analysis6m 10s
-
(Locked)
Interpreting a perceptual map3m 14s
-
(Locked)
Using cluster analysis and decision trees together9m 29s
-
(Locked)
A BIRCH/two-step example5m 12s
-
(Locked)
A self organizing map example7m 43s
-
(Locked)
-
-
-