TETIS - Text-Mining Team Repository Guide

This repository serves as the central hub for software production by the Text-Mining team at TETIS. It includes tools developed by our team, code for reproducibility, internship projects, and training materials.

---

🛠 Tools

Type	Name	Description	Repository Link
Local Python Package	SNEtoolkit	Absolute Spatial Named Entities extraction and disambiguation tool	GitHub
Local Streamlit Application	THECOB	Automatic protocol for the constitution of spatio-temporal and thematic corpora	GitHub
local Streamlit Application	GeospaCy	GeospaCy is a web application built in Python language used for extracting spatial relation entities (spatRE) from text and Geo-referenced them	Github
python script	CompEBS	This project aims to compare several EBS tools in terms of spatio-temporal and thematic aspects	institutional gitlab
Pip Package	GeoNLPlify	NLP library for data augmentation focusing on spatial information contained in text	GitHub

---

📄 Article Reproducibility

Journal/Conference	Name	Description	Repository Link
AGILE-2021	H-TFIDF	This project aims to extract discriminative terms on spatial and time windows	institutional gitlab
LREC-2022	Enriching Epidemiological Thematic Features For Disease Surveillance Corpora Classification	-	Github
ESWA	Explainable epidemiological thematic features for event based disease surveillance	-	Github
DS-2024	Geographical Biases in LLMs	Evaluation of the quality of LLM geo knowledge	GitHub
EGC-2025	Text 2 SQL for LandMatrix	Text-to-GraphQL / API REST for querying the Land Matrix database	GitHub

---

🎓 Work of our students

Author	Name	Description	Repository Link	Year
Aicha Zouhair	landmatrix-resourcecontracts-feeder	Export mining deals from ResourceContracts to Land Matrix.	github	2025
Fatiha Ait Kbir	Text 2 SQL for LandMatrix	Text-to-GraphQL / API REST for querying the Land Matrix database	GitHub	2024
Nelson Jaimes-Quintero	food-insecurity-risk-mining	Automatic named entity recognition pipeline to identify possible drivers of food insecurity in French-language news articles. The project supports event extraction (EE) using sentiment analysis and links TIME and LOCATION entities to event mentions.	GitHub	2024

---

📚 Training Materials

Type	Name	Description	Link
Google Colab	2022 - H2020 MOOD PhD school	This notebok is used for the "Mining Media Data" session of the MOOD Summer School 2022	GitHub
Google Colab	2023 - pratical-session-nlp-for-one-health-murdoch-mood	Practical session on NLP for One Health - Murdoch Mood	GitHub
Google Colab	2024 - Geographical Biases in LLMs	Evaluation of the quality of LLM geo knowledge	GitHub
Google Colab	2024 - ETTM INRAE/DipSO/ASTRA: Vectorisation, clusterisation and classification		institutional gitlab

---

Feel free to explore, contribute, and reach out with any questions! 🚀