Unsupervised Machine Learning for Cancer Epidemiology Mapping in New Caledonia

This project applies unsupervised learning to analyze cancer epidemiology in New Caledonia. It aims to discover hidden patterns and at-risk population groups using clustering algorithms. The local context presents unique challenges: significant geographical disparities in healthcare access, specific environmental risk factors, and cultural diversity affecting health behaviors. This data-driven mapping seeks to support public health strategy optimization across the Caledonian territory, helping to target prevention and resources more effectively.

Dataset

This project utilizes the "Épidémiologie descriptive des cancers en Nouvelle-Calédonie" (Descriptive Epidemiology of Cancers in New Caledonia) dataset, which was obtained from the official open data portal of New Caledonia: data.gouv.nc

Technologies used

• Langage : R

Installation

Clone the repository

git clone https://github.com/naima-beck/machine_learning_cancer_new_caledonia.git
cd machine_learning_cancer_new_caledonia

Structure of the projet

machine_learning_cancer_new_caledonia/
├── data/
   ├── raw/
   └── processed/           
├── report.pdf
├── notebook/
   ├── 1_Cleaning.Rmd
   └── 2_Visualisation.Rmd           
└── README.md

Authors

• @naima-beck
• @axellelepoul-ctrl

Cy Tech - Sciences-Po Saint-Germain-En-Laye - [2025/2026]

License

This project is carried out in an academic setting. The data is provided for educational purposes. It is licensed under CC BY-NC-SA 4.0.