Skip to content

A now defunct website that was built to provide insights into data regarding a user's favorite songs/albums by visualizing data from the Spotify API as well as lyrical information derived from supervised learning ML algorithms.

License

Notifications You must be signed in to change notification settings

noahari/MusicAnalytics

Repository files navigation

THIS PROJECT IS DEPRECATED

This project is no longer receiving updates and may have SIGNIFICANTLY out of date dependencies

MusicAnalytics

What for?

This website allows the users to input a list of songs and albums by their favorite artists, and in turn outputs an interactive parallel coordinates graph. The user can then select different lines to focus on specific data points, and selectively remove data that is less interesting to them. There is also the BANGATRON3000 component. Though it is currently disabled for frontend development, its backend implementation is complete and uses the same data points fed into a machine learning algorithm to determine whether a song is a party banger or not. This was left intentionally unpersonalizable to comply with the terms of the APIs used in this project.

Getting Started

Simply access the website, add songs or albums to the queue, and then push the queue to the graph.

Prerequisites/Pips Used

requests
pandas
numpy
re
textstat
scikit

Features

ML NLP features developed using skl:
Sentiment Analysis
Lexical Analysis
Writing Level

Ideas to Implement

Single Song

Sentiment Analysis
Word Frequency
Thematic Analysis
Writing Level (lmao)
Allusions/References
Scansion
GETTERS

Long Term Ideas

Mapping over multiple songs for Week/Month/Year Analysis
Supervised Learning (to predict wanted?)
Advanced Data
Song Structure
Sentiment Analyser Supervised Machine Learning Upgrade

About

A now defunct website that was built to provide insights into data regarding a user's favorite songs/albums by visualizing data from the Spotify API as well as lyrical information derived from supervised learning ML algorithms.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published