This repo contains my personal portfolio for the class computational musicology.

The corpus is the dataset of music I will analyze over the coming weeks. I decided to analyse my top songs of the past years, which spotify yearly curates every year for its "Wrapped 20XX" feature. On top of these "top of the year" playlists, I decided to include the "My Time Capsule" playlist. This playlist is generated by spotify based on your taste in music and is supposed to result in a personal throwback mix. Therefore my corpus consists of:

• YourTopSongs2021
• YourTopSongs2020
• YourTopSongs2019
• YourTopSongs2018
• YourTopSongs2017
• YourTopSongs2016
• YourTimeCapsule

This dataset consists of 650 songs over a span of six years. I'm interested if and how my taste and music has changed, if there are trends which could correlate with events in my life, and if I can uncover the recipe for the "Time Capsule" Playlist. How does the distribution between genres shift? Are there artists that stay consistently throughout the years, are there "rising stars"?

A strength of this dataset is that it should be expressive - I can't find the exact number right now, but as far as I remember my "minutes of listening" per year are around 50.000 minutes/year. A weakness (or maybe visible trend?) could be that over the past two years I've shifted away from listening on Spotify exclusively to using SoundCloud specifically for listening to Techno & House.

As (a-)typical songs I have selected the first and last song of the six yearly playlists: