This project analyzes human cognitive states using EEG signals while listening to different types of music.
Machine Learning and Deep Learning models are used to classify relaxation-related brain states.
https://drive.google.com/file/d/1nVckQGYF5HqQsnc39vJtyux4iYIButVE/view?usp=sharing
Music has a strong influence on human emotions and mental states.
This project uses real-time EEG signals to identify cognitive states such as Relaxed, Neutral, and Normal while listening to music.
The EEG signals are:
- Filtered using Bandpass Butterworth Filter
- Processed to extract Power & Statistical Features
- Classified using ML and DL models
- Collect real EEG signals during music listening
- Extract meaningful brainwave features
- Classify cognitive states using ML & DL models
- Compare model performance using accuracy and confusion matrix
- Alpha Music (Relaxation)
- Flute Music (Deep Calmness)
- Normal State (No Music)
- EEG Data Collection
- Signal Preprocessing (Bandpass Filter)
- Feature Extraction
- Classification (ML/DL Models)
- Performance Evaluation
- Python
- React.js
- Flask / FastAPI
- NumPy
- Pandas
- SciPy
- Scikit-learn
- XGBoost
- TensorFlow / Keras
- Matplotlib
- Random Forest Classifier (RFC)
- XGBoost
- LSTM (Long Short-Term Memory)
- Delta Power
- Theta Power
- Alpha Power
- Beta Power
- Gamma Power
- Mean
- Variance
- Standard Deviation
- Skewness
- Kurtosis
- Entropy
- Accuracy
- Precision
- Recall
- F1-score
- Confusion Matrix
pip install -r requirements.txt
python app.pynpm install
npm startpython app.pyOpen http://127.0.0.1:5000/predict
EEG-Cognitive-State-Analysis/
│
├── backend/
│ ├── preprocessing.py
│ ├── feature_extraction.py
│ ├── models/
│ ├── app.py
│
├── frontend/
│ ├── src/
│ ├── components/
│ └── App.jsx
│
├── dataset/
├── results/
├── README.md
└── requirements.txt