SFU CMPT 340 Project - EMOTIC - EEG Monitoring of Thoughts and Individual Conditions

A project using EEG to analyze and interpret users' emotional states through neural signal processing and machine learning.

• The collection of EEG data accredited to the authors of A Large Finer-grained Affective Computing EEG Dataset: https://rdcu.be/d1TNW
• Chen, J., Wang, X., Huang, C. et al. A Large Finer-grained Affective Computing EEG Dataset. Sci Data 10, 740 (2023). https://doi.org/10.1038/s41597-023-02650-w

Team Members

1. Maisha Supritee Chowdhury
2. Edan Stasiuk
3. Rashed Hadi
4. Zihao Xie
5. Daniel Kim

Important Links

Timesheet	Slack channel	Project report	Figma Designs

Table of Contents

1. Demo

2. Installation

3. Reproducing this project

1. Demo video

What to find where

repository
├── backend                      ## Flask backend
    ├── app.py                   ## server
    ├── model.pkl                ## trained svm model
    ├── requirements.txt         ## required packages for project
├── frontend                     ## NextJS frontend
├── README.md                    ## You are here
├── GRU_model                    ## Gated Recurrent Unit model 
    ├── model.py                 ## model
    ├── process.py               ## processing data for model
    ├── predict.py               ## trying to predict with this model
├── KNN_model                    ## K Nearest Neighbors model
    ├── knn_model.py             ## model
    ├── process_data.py          ## processing data for model    
├── process_data.py              ## process data for main svm model
├── svm_model.py                 ## main model used for prediction
├── generate_sample.py           ## generates 1 sample for each emotion to try to predict

2. Installation

Clone the Repository

git clone https://github.com/sfu-cmpt340/EMOTIC_g3.git
cd EMOTIC_g3

Install Dependencies

Frontend (Next.js)

cd frontend
npm install
npm i react-icons

Then navigate back to the main directory:

cd ..

Backend (Flask)

cd backend
python3 -m venv .venv
. .venv/bin/activate
pip install -r requirements.txt

3. Reproduction

(Optional) Running the Web Application and Server

To run the full-stack application locally, you need to start both the backend server and the frontend web application. Follow the steps below:

1. Start the Backend Server

Navigate to the backend folder and run the Flask server with your virtual environment activated:

cd backend
flask --app app run

2. Start the Frontend

Open a new terminal, navigate to the frontend folder, and start the Next.js development server:

cd frontend
npm run dev

Open your browser and navigate to http://localhost:3000/ to use the web application. Instructions to download data to use with the application can be found below.

Running the ML Model Standalone

1. Create the venv and install packages from the main directory:

python3 -m venv .venv
. .venv/bin/activate
pip install flask flask_cors pandas matplotlib seaborn scikit-learn tensorflow numpy

2. Download sample data and run

wget --no-check-certificate 'https://docs.google.com/uc?export=download&id=1DQPpjJIigouQEzcrkb9SQbCOPtSVbP5U' -O sample_data.zip
unzip sample_data.zip
python3 svm_model.py
>> Enter the name of the sample file: filename

filename should match any of the 9 csv files in Sample_data

Troubleshooting: wget does not work

Data can be found at https://drive.google.com/file/d/1DQPpjJIigouQEzcrkb9SQbCOPtSVbP5U/view?usp=sharing

• Download and unzip in the root directory of this project
• There should now exist a folder called Sample_data with 9 csv files

3. Output

Running the above python command should now display the predicted emotion of the selected csv data in the terminal