MoodSense-

KO | EN

gitlite — search

#typescript #ai-agents #ai #dsh-plugin #deepseek-harness #open-source #claude-code #codex #cli #developer-tools #react #windows

★ 63 Open GitHub ↗

Music Recommendation Based on Mood

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

Mood Sense-ml Readme
🎵 MoodSense – Music Recommendation Using Machine Learning

Predict music mood categories using audio features with a fully trained ML model.

MoodSense is a simple, clean, and beginner-friendly machine learning project that classifies music into mood categories such as Happy, Sad, Calm, and Angry based on audio features like energy, tempo, valence, and danceability.

This project is designed to be:

✅ Easy to run

✅ Fully trainable

✅ Lightweight (no GPU required)

✅ GitHub-friendly

✅ Perfect for ML beginners & portfolios

🚀 Features

Supervised Machine Learning (K-Nearest Neighbors)

Dataset included

Model training script

Prediction script

Clean project structure

Easy to extend to Spotify API

📂 Project Structure
MoodSense-ML/
│
├── data/
│   └── mood_music.csv
│
├── src/
│   ├── train.py
│   └── predict.py
│
├── model/
│   └── mood_model.pkl  (auto-generated after training)
│
├── requirements.txt
└── README.md
📊 Dataset Description

The dataset contains music audio features and a mood label.

Feature	Description
energy	Intensity of the song (0–1)
tempo	Speed of the song (BPM)
valence	Positivity score (0–1)
danceability	How suitable for dancing (0–1)
loudness	Volume level (negative dB)
mood	Target label (Happy, Sad, Calm, Angry)

The dataset is synthetic for demonstration purposes and can be extended using Spotify audio features.

🧠 Machine Learning Model

Algorithm Used: K-Nearest Neighbors (KNN)

Why KNN?

Works well for similarity-based problems

Easy to understand

Lightweight and efficient

Perfect for educational and portfolio projects

⚙️ Installation & Setup
1️⃣ Clone the repository
git clone https://github.com/Sindu0706/MoodSense-ML.git
cd MoodSense-ML
2️⃣ Install dependencies
pip install -r requirements.txt
🏋️ Train the Model

Run:

python src/train.py

Output:

Model accuracy printed

model/mood_model.pkl file created

🔮 Make Predictions

Run:

python src/predict.py

Example Output:

🎧 Recommended Mood: Happy
📈 Example Use Cases

Music streaming recommendation systems

Mood-based playlist generation

Emotion-aware AI assistants

ML educational demonstrations

🛠️ Future Improvements

Integrate Spotify API for real audio features

Deploy as Flask web application

Convert into mobile app backend

Replace KNN with Deep Learning model

Add real-time emotion detection

⚖️ Ethical Considerations

No personal user data collected

No emotion manipulation

Dataset is synthetic for demonstration

⭐ If You Like This Project

Give it a ⭐ on GitHub and feel free to contribute!

👩‍💻 Author

Developed as a Machine Learning portfolio project.

MoodSense-

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README