Clustering
Clustering requires a free account
Takes 10 seconds, no payment required. Here's what you get:
- Unlocks Clustering and 7 other advanced tools
- Higher file processing limits, no daily cap
- Save and re-run workflows across devices
About Clustering
Group similar data points automatically.
This tool applies the K-Means algorithm to your numeric columns, iteratively assigning each row to the nearest of K cluster centroids and recalculating those centroids until the groups stabilize.
It's a form of unsupervised learning: there's no target column to predict, only structure to discover in how rows sit relative to each other across your chosen features.
Because K-Means is distance-based, features on wildly different scales (like revenue in thousands next to a 0-1 rate) will distort the grouping unless normalized first.
Category: AI & Machine LearningCommon Use Cases
- Segmenting customers by purchase behavior for targeted campaigns
- Grouping sensor or IoT readings into normal operating states
- Finding natural product tiers from price and usage metrics
Key Features
- K-Means Algorithm
- Configurable Number of Clusters (K)
- Multi-Column Numeric Grouping
- Cluster Label Export
Run Scaling first so every feature contributes proportionally to distance calculations, then use Clustering to assign group labels you can chart or filter on afterward.
Frequently Asked Questions
It groups similar data points together automatically based on their numeric feature similarity.
You control the clustering parameters, including how many groups to look for.
A practical starting point for analysts doing customer segmentation or pattern discovery without setting up a full Python ML environment.
Read the related guide →