This repository contains Jupyter Notebooks that demonstrate geospatial clustering techniques applied to urban datasets. The notebooks utilize popular Python libraries such as pandas, geopandas, scikit-learn, and matplotlib to perform clustering analysis and visualize the results.
-
01_Educational_Hubs.ipynb: A basic application that uses K-Means and DBSCAN clustering to characterize the spatial distribution of educational institutions in Porto, Portugal. -
02_Wildfire_Hotspots_in_Portugal.ipynb: An advanced application that employs HDBSCAN clustering to identify and analyze wildfire hotspots across Portugal. -
03_Spatiotemporal_Vulnerability.ipynb: A end-to-end application based on VERUS, which combines OPTICS and K-Means clustering to assess spatiotemporal vulnerability in urban areas based on various points of temporal influence.
You can run the notebooks directly in Google Colab without any local installation. Simply click the "Open in Colab" badge at the top of each notebook.
⚠️ Note: Colab provides a temporary environment. Your work will be lost after your session ends, so remember to download any important data or notebooks.
To run the notebooks locally, follow these steps:
-
Clone the repository:
git clone https://github.com/les2feup/geo-clustering.git cd geo-clustering -
Create a virtual environment and activate it:
python3 -m venv venv source venv/bin/activate # On Windows use `venv\Scripts\activate`
-
Install the required dependencies:
pip install -r requirements.txt
-
Open the repository in VSCode and launch Jupyter Notebooks.
Python 3.12 is the tested version for running the notebooks locally. The complete
dependency specification is maintained in requirements.txt.
- Notebook runtime:
ipykernel,requests - Geospatial analysis:
geopandas,osmnx,verus - Clustering and machine learning:
scikit-learn,hdbscan - Visualization:
matplotlib,contextily
This project is licensed under the Apache 2.0 License. See the LICENSE file for details.
- This project was developed at the Laboratory of Emerging Smart Systems (LES2) at the Faculty of Engineering of University of Porto (FEUP), Portugal.
- This work was supported by the Associate Laboratory Advanced Production and Intelligent Systems – ARISE LA/P/0112/2020 (DOI 10.54499/LA/P/0112/2020), by the Base Funding (UIDB/00147/2020) and Programmatic Funding (UIDP/00147/2020) of the R&D Unit Center for Systems and Technologies -- SYSTEC, and by the Fundação para a Ciência e a Tecnologia (FCT) through the PhD scholarship (2024.02446.BD).