Robust source tracing for solid waste via machine learning-enabled mineralogical fingerprinting

The work presents a novel mineralogical fingerprinting approach to accurately identify the source of heavy-metal hazardous solid wastes using machine learning.

---

🌟 Features

• 🔍 Interactive Web App: Input mineral phases and predict waste source in real time.
• 📊 Visual Analytics: Interactive probability bar charts and 2D similarity visualization via MDS.

---

🚀 Quick Start

Prerequisites

• Python ≥ 3.8
• pip or conda

Installation

# Clone the repository
git clone https://github.com/Laaery/SWFP.git
cd FP_HMHSW

# Install dependencies
pip install -r requirements.txt

📝 Citation

If you use this code or framework in your research, please cite our paper:

@article{LIN2026109068,
title = {Robust source tracing for solid waste via machine learning-enabled mineralogical fingerprinting},
journal = {Resources, Conservation and Recycling},
volume = {234},
pages = {109068},
year = {2026},
issn = {0921-3449},
doi = {https://doi.org/10.1016/j.resconrec.2026.109068},
url = {https://www.sciencedirect.com/science/article/pii/S0921344926002922},
author = {Le Lin and Changhai Ren and Shen Qu and Yang Xiao and Yin Li and Xueming Liu and Han Wang and Zhang Lin}
}