|
|
---
|
|
|
license: mit
|
|
|
tags:
|
|
|
- recommendation
|
|
|
- binary-classification
|
|
|
- music
|
|
|
- ensemble
|
|
|
---
|
|
|
|
|
|
# 🎵 Music Recommendation System
|
|
|
|
|
|
This repository hosts a collection of machine learning models designed to recommend songs by predicting whether a user is likely to "like" a track based on its audio features.
|
|
|
|
|
|
## 📁 Files Included
|
|
|
|
|
|
- `data.csv` — Dataset of 195 songs with features like danceability, energy, loudness, tempo, etc.
|
|
|
- Trained model files:
|
|
|
- `logistic_regression.joblib`
|
|
|
- `random_forest.joblib`
|
|
|
- `xgboost.joblib`
|
|
|
- `svm.joblib`
|
|
|
- `voting_classifier.joblib`
|
|
|
- `catboost_model.cbm`
|
|
|
- `ann_model.keras`
|
|
|
- `final_model_card_scaled.pdf` — Full model evaluation, comparison table, and chart
|
|
|
|
|
|
## 🧠 Models Used
|
|
|
|
|
|
- Logistic Regression
|
|
|
- Random Forest
|
|
|
- XGBoost
|
|
|
- Support Vector Machine (SVM)
|
|
|
- Voting Classifier (Ensemble)
|
|
|
- CatBoost
|
|
|
- Artificial Neural Network (ANN)
|
|
|
|
|
|
## 📊 Evaluation
|
|
|
|
|
|
All models were evaluated using:
|
|
|
- Accuracy
|
|
|
- Precision
|
|
|
- Recall
|
|
|
- F1-Score
|
|
|
|
|
|
Refer to the PDF `final_model_card_scaled.pdf` for full details.
|
|
|
|
|
|
## 📬 Contact
|
|
|
|
|
|
Maintained by Sujal Thakkar.
|
|
|
|
