CRM Experiment

This project implements a Continuous Self-Reporting Method (CRM) experiment to collect two-dimensional emotional data ( 2DES) from participants (Schubert, 2010). The project is divided into three services: a WebSocket server, a Next.js frontend, and a Flask backend. It allows synchronized data collection from multiple participants through WebSockets.

Architecture

The project consists of three main components:

• WebSocket Server: Manages real-time communication between the frontend and backend, sending start and stop signals to participants.
• Next.js Frontend: Provides the user interface for participants and the admin, displaying the experiment and results pages.
• Flask Backend: Processes and stores data, resampling and synchronizing collected data, and storing it in an SQLite database. Generates plots using Matplotlib for the results page.

Workflow

Participant Interaction:

1. Participants open the webpage.
2. Admin initiates the experiment.
3. Audio stimulus plays only on the admin's computer.
4. WebSocket server sends start signal to participants.

Data Collection:

1. Participants' emotional data is recorded continuously.
2. Data is resampled and synchronized.
3. Data Storage and Analysis:

Data is stored in an SQLite database. Results can be visualized later on the results page.

Usage

To run the project, you need Docker and Docker Compose installed on your machine. Then, you can start the services with:

docker-compose up

This will start the WebSocket server, frontend, and backend. The frontend will be available at http://localhost:80. To start the live recording of data, navigate to http://localhost:80/admin.

License

This project is licensed under the MIT License - see the LICENSE file for details.