Advanced Analytics · TOOL GUIDE

K-Means Clustering

Open Quantly ↗

Understand the method. Use it with confidence.

Partition data into k distinct clusters. Quantly runs the method against your active analysis context and returns structured results for review and interpretation.

01

When to use it

Use this tool when the question requires multivariate structure, segmentation, simulation, causal-design assumptions, or model explainability.
Use it only when the selected variables and study design match the method's requirements.

02

How to use it

  1. Open the tool from the Quantly Tools panel.
  2. Confirm the active dataset and analysis session.
  3. Set Select Features (Required): Choose numeric and/or categorical columns for clustering.
  4. Set Number of Clusters (k) (Optional): The number of clusters to form. Default is 3.
  5. Set Auto-find k (Elbow Method) (Optional): Automatically suggest optimal k using the Elbow Method.
  6. Set Categorical Encoding (Required): How to transform text variables for the model. One-Hot is usually safer for this model.
  7. Set Save as New Dataset (Optional): Create a new dataset with the assigned Cluster labels.
  8. Review optional or advanced settings and keep defaults unless your design requires a change.
  9. Run the tool, inspect warnings and diagnostics, then save or export the result with its assumptions.
03

How to interpret results

Interpret results in the context of preprocessing, tuning choices, stability, validation, and domain knowledge. Complex outputs are sensitive to assumptions and should be checked with alternative specifications.