Back to Tools

Semantic Keyword Clustering

Use cases

Planning content hubs and pillar pages Deciding which keywords should share a page vs have separate pages Building taxonomy structures for eCommerce Organising keyword research for content teams

Groups keywords by meaning using SentenceTransformer embeddings (all-MiniLM-L6-v2 model) and PolyFuzz clustering.

Configurable similarity threshold (default 0.80), optional Porter stemming, and volume-based hub naming.

Outputs interactive sunburst/treemap visualisations.

Platform

Python script (requires Python 3.x)

Input

Keywords CSV (optionally with search volume)

Output

CSV with clusters and interactive HTML chart

View Source

Features

  • SentenceTransformer embeddings (all-MiniLM-L6-v2)
  • PolyFuzz clustering with configurable similarity threshold
  • Volume-based cluster naming (highest volume = hub)
  • Interactive HTML visualisations (sunburst/treemap)
  • CPU or GPU acceleration support

How to use

  1. 1 Prepare your keyword list in CSV format
  2. 2 Run the CLI tool with your file path
  3. 3 Configure similarity threshold if needed (default 0.80)
  4. 4 Review the generated clusters (hub = primary keyword, spokes = related terms)
  5. 5 Explore the interactive HTML visualisation
  6. 6 Export clustered keywords as CSV or Excel

Let's work together

Monthly retainers or one-off projects. No lengthy reports that sit in a drawer.

Let's Talk