Skip to content

tansawit/rap-artist-classifier

master
Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.

Rap Artist Lyrics Classifier

Done as part of the final project for the University of Michigan's EECS 486 class.

The aim of this project to explore the linguistically unique nature of rap lyrics and the effectiveness of various classification models in matching those lyrics to the respective artists. The project uses a popularity-based seedlist of 44 rappers, along with information and lyrics of their respective top 50 songs, requested from Genius through its API. We then train the various classification models, using a 80-20 train-test split, on different combinations of the data and evaluated according to the recall metrics to determine the effectiveness of each combination.

The web app then allows users to input their own lyrics or strings of text and the preferred classification model, then outputs the rap artist with the closest matching lyrics features.

A detailed summary of the project, methodologies, evaluation, and results can be found in the PDF report.

Contributors: Thomas Horak, Brian Guo, Krishan Amin, Sawit Trisirisatayawong, Vraj Desai

About

Paper examining the effectiveness of various classification models in matching rap song lyrics to corresponding artists

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages