From the course: Complete Guide to Generative AI for Data Analysis and Data Science

Unlock the full course today

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

Clustering with DBSCAN

Clustering with DBSCAN

From the course: Complete Guide to Generative AI for Data Analysis and Data Science

Start my 1-month free trial Buy for my team

Clustering with DBSCAN

“

- [Instructor] We had some decent results using K-means clustering in our previous video with our customer segmentation. But now I want to try a different clustering algorithm. I want to modify the script to use DBSCAN which is density based scanning algorithm instead of k-means. And let's see if we can improve on that or if this actually takes us to a poorer example. So looks like it tried to use DBSCAN and they ran into some problems initially. Oh so our Silhouette score is better it's .84. So the number of clusters that was created this time was six and the Silhouette score was 84 so we're getting better, we're getting closer to one here. So let's take a look at the code and see what happened here. So it looks like we've got errors we were building the initial script generated an error, a raise value error it looks like it didn't meet some minimum requirements. Oh, interesting it appears the DBSCAN algorithm identified all points as noise so they're adjusting a couple of the model…

Contents