From the course: Machine Learning and AI Foundations: Clustering and Association

Unlock the full course today

Join today to access over 24,800 courses taught by industry experts.

Finding optimum value for k: k = 4

Finding optimum value for k: k = 4

From the course: Machine Learning and AI Foundations: Clustering and Association

Start my 1-month free trial Buy for my team

Finding optimum value for k: k = 4

“

- [Instructor] Trial and error is the name of the game everybody, so we're going to go ahead and try K=4. Now, with K means, the way it's set up here in modeler each time I run this, I would get the same results. So I'm not saving different versions of this model I'm just changing the value of K. Obviously, some folks might choose to do that and have more than one diamond on the stream, I'm just not doing so right now. In a real-world project, I probably would, especially if there were lots of variations on the theme. Like if I was sending all the cases to one model, fewer than all the cases to another model, different numbers of variables and so on. But here, I'm keeping everything the same except for that value of K. Let's take a look. And again, what I usually turn to first is highlighting all the clusters. Now remember, in your tool of choice, you might not have this, you know, fancy cluster viewer. But you're going to have a way of doing box-and-whiskers plots so you can…

Contents