Through its API, Spotify has a wealth of data linked to individual users making it a rich database for data and music lovers alike to explore their personal music. Using the Spotipy package I do some exploratory analysis on my own music data. As an avid user of the platform since 2013, and having changed accounts in 2015 for the prolongation of the student discount, I have almost 10 years of data to look at.
Here is an EDA of my music taste, with some possbible research questions to pursue:
- Link to Slide Deck
- Jupyter Notebook in repository
The question I am most interested in exploring is why there was an increase in valence in 2022. Is this due to a shift in genre, key, or maybe mode?
Here you can see that 2022 was the only year that had exclusively pop genres in its top 5
Generally, the genres from 2022 had a higher valence than did the genres of 2016 (which had the lowest overall valence than any other year)
| 2022 | 2016 |
|
|
A limitation of this study, however, is the fact that songs do not have individual genres in the API--only artists do. The genres for each song where the first genre associated with an artist.