This repository includes the code and parts of the data used in the article titled "Text Mining Tafsir: Compilation and Preliminary Explorations of a Curated Corpus of 80 Qur’anic Commentaries," published in the Journal of Digital Islamicate Research. Please note that the Jupyter notebooks in this repository are solely intended to document the clustering process. The individual steps can only be replicated with the necessary data, including the Mistral AI embeddings. The data originates from Adrian Bernhard's Tafsir database, available on Zenodo. This repository includes: The Jupyter notebook Clustering.ipynb used to retrieve embeddings and conduct the clustering analysis. Most of the data, tables, images, and csv files used in the article in the respective subfolders. Smaller parts of the EDA in the Jupyter notebook EDA.ipynb. All embeddings retrieved via the Mistral API (in the form of pickle files) which build the foundation of the clustering analysis. The SQL queries used in the database to create certain statistics (see the file SQL_query.md).