Wrangling and Visualizing Musical Data
Wrangle and visualize musical data to find common chords and compare the styles of different artists.
- 10 tasks
- 1,636 participants
- 1,500 XP
Project Description
Apply data-wrangling and visualization tools from the tidyverse to musical
data. Find the most common chords and chord progressions in a sample of
pop/rock music from the 1950s-1990s, and compare the styles of different
artists. This project assumes familiarity with standard TidyVerse tools
for R, in particular the tibble data structure and the dplyr and
ggplot2 packages. No specific musical knowledge is required, though it
may give you ideas for further exploration of the dataset after completing
the project.
Before taking on this project we recommend that you have completed the following courses:
This project uses a parsed and cleaned version of the McGill Billboard Dataset, version 2.0 (CC0 license).
Project Tasks
- 1 Introduction
- 2 The most common chords
- 3 Visualizing the most common chords
- 4 Chord "bigrams"
- 5 Visualizing the most common chord progressions
- 6 Finding the most common artists
- 7 Tagging the corpus
- 8 Comparing chords in piano-driven and guitar-driven songs
- 9 Comparing chord bigrams in piano-driven and guitar-driven songs
- 10 Conclusion
Kris Shaffer
Data scientist and instructional technology specialist
Kris Shaffer, Ph.D., is a data scientist and instructional technology specialist at the University of Mary Washington. He also does freelance work in web intelligence and analytics, and has an academic background in music theory and the digital humanities. You can find him on the web at pushpullfork.com.
See More