From the course: SPSS: Data Visualizing and Data Wrangling

Unlock the full course today

Join today to access over 24,600 courses taught by industry experts.

k-means clustering

k-means clustering

From the course: SPSS: Data Visualizing and Data Wrangling

Start my 1-month free trial Buy for my team

k-means clustering

“

- [Instructor] Putting like with like is one of the most important tasks in the data world, finding out where the similarities are in your data. Now, in a previous video, we looked at hierarchical clustering, which lets you ask that as a very open-ended question, where you end up with one giant category that has every single observation with it down to every observation being in its own category. And you can look at all the in-between levels. A different approach, which is often used for similar purposes is called k-means clustering, where k is a number of clusters, you can pick two, three, five, whatever, and it's based on means or centroids in a multidimensional space. You can also use it to see how the data stacks up against existing categories. And that's actually what I want to demonstrate right here. To do this, we're going to use our estate data set and let's come up here and go to analyze down to classify. But instead of doing hierarchical cluster, we're going to do this one…

Contents