NEKO

You have the vectors. You do not know how many groups exist.

k-means asks for k before you look at the data. Density methods ask for parameters. Either way the number of groups is a decision you made — not a result you can defend.

Doing it yourself means writing the pipeline: distances, minimum spanning tree, percolation, the cut, the optimization. Weeks of an expert’s time, a result nobody else can reproduce, and the same arbitrary number at the end. Create your account and let the platform find the groups in your data.

Analysis pipeline

Projects → Datasets → Inferences → Reports

  1. Project
  2. Dataset
  3. Inference
  4. Report

Let's build a pipeline 🚀 Do you want to create a new project or use an existing one?

Demo dataset

politics.csv

421 rows · 549 columns

Create an account and run your own dataset.

The graph above came from the same pipeline that runs for a project: your dataset goes in, the result comes out ready to open and share.

Create account

Already have an account? Sign in