APRIL Project

Analysis of Coastal Risk Perception in Occitanie (A.P.Ri.L: Analyse de la Perception des Risques Littoraux en Occitanie)

The APRIL project analyzes how the population of the Occitanie region of southern France perceive coastal risks such as sea-level rise, erosion, and extreme weather events. By collecting and analyzing diverse web-based data (news articles, reports, social media, etc.), the project aims to inform adaptive coastal management policies through insights gained from Natural Language Processing (NLP).

Installation

1. Clone the repository:

git clone https://github.com/thibaut-dst/APRIL.git

2. Build the Docker image:

docker build -t april-web:1.1 -f build/Dockerfile .

Or pull the pre-built image:

docker pull registry.mde.epf.fr/april-web:1.1

3. Ensure the docker-compose.yml file uses the correct version (1.1).

Start the application:

cd build
docker compose up -d

Prerequisites

• Python 3.12+
• pip (Python package installer)
• Docker & Docker Compose

Usage

Once docker containers are runing, access the application at:

http://127.0.0.1:5000/

1️⃣ Launch the Data Collection Pipeline

Navigate to http://127.0.0.1:5000/launch-pipeline to access the pipeline controls. Start the web scraping process by clicking "Start Pipeline". You can stop the process anytime with the "Stop Pipeline" button or let it automatically complete all search combinations.

2️⃣ Run the Text Processing Pipeline

After data collection finishes, click "RUN NLP" to start the text processing pipeline. This process will automatically complete without further action.

3️⃣ Browse and Filter Documents

Return to the main page to view the list of documents in the database. Use the filters to refine your search and click on any document to view its content and metadata.

🔄 Need More Documents?

Repeat Steps 1 and 2 to collect and process additional data.

For more details, refer to the project documentation.

Main Features

• Data Collection: Automated scraping of web-based data sources related to coastal risks.
• NLP Analysis: Identification of key themes, Named Entity Recognition, semantic scoring.
• Interactive Dashboard: Visualization of data insights to support policy-making.

Explore more in the Wiki 'Features'.

Authors

• thibaut-dst: Spearheaded the design, system architecture, NLP feature development, frontend and UI/UX design.

• l-gou: Led the project, NLP features, managed documentation, and handled testing efforts.

• theoP17: Worked on frontend development and conducted research on available options and best practices.

• antoinebtb: Focused on backend development and API development.

• pharaoph09: Contributed to backend development and wrote documentation.

License

MIT License © 2024 APRIL Project Team