https://github.com/brendanddev/music-vector-recommender
https://github.com/brendanddev/music-vector-recommender
Last synced: 10 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/brendanddev/music-vector-recommender
- Owner: brendanddev
- Created: 2025-07-30T23:18:14.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2025-07-30T23:48:12.000Z (11 months ago)
- Last Synced: 2025-07-31T02:07:17.498Z (11 months ago)
- Language: Python
- Size: 1000 Bytes
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Music Vector Recommender
A lightweight command-line music recommendation tool that uses song lyrics to suggest similar tracks based on vectorized semantic similarity.
---
## Features
- Recommends songs based on lyric similarity using TF-IDF vectorization.
- Cosine similarity determines how closely songs match.
- Removes duplicate song titles for clean suggestions.
- Easy to configure number of recommendations, dataset location, etc.
- Optional debug mode to view raw similarity scores for transparency and testing.
---
## How It Works
1. Lyrics are cleaned and vectorized using TF-IDF.
2. The selected song's lyrics are compared against all others using cosine similarity.
3. The most similar (non-duplicate) songs are recommended.
---
## Dataset
You can provide your own dataset in CSV format (title, artist, lyrics), or use the sample below.
---
## Attributions
- [Spotify Web API](https://developer.spotify.com/documentation/web-api/)
- [Kaggle Genius Lyrics Dataset](https://www.kaggle.com/datasets/carlosgdcj genius-song-lyrics-with-language-information)