Clustering Benchmark Datasets


The Clustering Benchmark Datasets contain 3 synthetic datasets conforming a structured domain with intuitively separable clusters in different forms as shown below.
Blobs Dataset
notion image
Circles Dataset
notion image
Moons Dataset
notion image
All three datasets include 500 data points in 2-dimensional space.
This dataset is often used for experimenting with clustering techniques and exploring their performance visually.

Data Location

Storage → Samples → segmentation_blobs.csv
Storage → Samples → segmentation_circles.csv
Storage → Samples → segmentation_moons.csv
Data Description
X-coordinate of the point
Y-coordinate of the point