Input: List of keywords (and optional search volume, difficulty).
Output: Cluster labels and optional cluster names (e.g. from centroid keywords).
Models: sentence-transformers embeddings + KMeans/HDBSCAN; optional UMAP for visualization.

Tech Stack

Category	Tools
NLP	sentence-transformers
Clustering	scikit-learn (KMeans, HDBSCAN)
Data	pandas, NumPy

Setup

pip install -r requirements.txt

Usage

bash

python train.py
python inference.py --input data/keywords.csv --output data/clustered.csv

Project structure

06_keyword-cluster-seo/
├── config.py
├── train.py           # Fit encoder + KMeans; save to models/
├── inference.py       # Assign cluster to keywords CSV
├── requirements.txt
├── .env.example
├── data/
│   └── keywords.csv      # Sample: one column "keyword"
└── models/

Data

Sample data (included): data/keywords.csv — single column keyword. Optional: volume, difficulty.
Set DATA_PATH and N_CLUSTERS in .env if needed.

License

MIT.

alirezaaminzadeh/keyword-cluster-seo

Keyword-Cluster-SEO: Intent-Based Keyword Clustering for Content Strategy

Author

Problem

Approach

Tech Stack

Setup

Usage

Project structure

Data

License

alirezaaminzadeh/
keyword-cluster-seo