briannogueira

Computational Analysis of the Greek New Testament

Researcher: Brian Nicolas Nogueira — Faculdade Batista do Rio de Janeiro Seminário do Sul ORCID: 0009-0004-0705-0710 Preprint DOI: 10.5281/zenodo.23695706

Summary

The lexical and morphological structure of the Greek New Testament is intensively studied, yet the underlying analyses are rarely reproducible. This project builds an open, agent-assisted pipeline that ingests public-domain Greek New Testament texts, normalises and tokenises them, reuses open morphological annotations, and computes reproducible lexical statistics across books and genres.

Methods

1. Corpus — assemble public-domain editions (SBLGNT; Robinson–Pierpont) with stable verse coordinates.
2. Normalização — Unicode NFC normalisation and Koine-tuned tokenisation.
3. Anotação morfológica — reuse MorphGNT tags; flag low-confidence tokens.
4. Estatística lexical — lemma frequency, hapax legomena, type–token ratio, genre distribution, collocations.
5. Pipeline agêntico — implemented in Google Antigravity, versioned and containerised for exact re-runs.

Deliverables

• Open-source pipeline (documented, tested)
• Citable dataset of lexical statistics (DOI)
• Preprint with method, results and limitations

Contact

• E-mail: brian.33842@seminariodosul.com.br
• ORCID: 0009-0004-0705-0710
• GitHub: https://github.com/briannogueira