Global Music Trends
Goal
Exploration of musical trends by year to year, differing by countries through employing statistical analysis of variables like genre, beats per minutes, length of songs, etc in order to generate conclusions on how music as a whole has transgressed and its potential trajectory going forth.
Datasets
Utilizing two datasets from Kaggle. Firstly - top10s.csv which includes the top songs from the 2010-2019 in the world by Spotify and Billboards. 13 variables are explored and original intent was to make inferences based upon genre, mean of minutes per songs, and yearly trends. Second - top50contry.csv which includes the top 50 Spotify songs by each country from 1942-2019. 13 variables are explored as well, and original intent was the same as the prior dataset.
Link to datasets:
Motivation
Music has been apart of humanity since the beginning of our creation. It plays a important role in countless lives for various reasons - whether for personal uses like: reducing stress, improving mood to more formal uses like: work songs, military marches, etc., the uses for music are countless. Due to its importance and prevalence in daily life globally - it’s important to understand the trends as humanity has progressed and be able to generate meaningful conclusions based on upon various factors that have influenced its transgression over the years.
Hope you enjoy!
Basic Overview
Top 10 Global Genres
| Top Genres | Song Count |
|---|---|
| adult standards | 145 |
| latin | 113 |
| dance pop | 76 |
| pop | 41 |
| album rock | 30 |
| desi pop | 30 |
| french hip hop | 22 |
| canadian pop | 21 |
| j-pop | 21 |
For clarity:
Pop Music - Also known as popular music can be classified as a genre which appeals to the taste of a large segment of a given population.
Adult Standards - Musical genre aimed at “mature” adults - those above the ages of 50 comprised of various genres.
Desi Pop/J-Pop - Cultural specific pop music; Desi is for the country of India and J is for the country of Japan.
Global - Pre & Post 2000 Top Genres
This is a illustration showing the progression of musical genres globally. Although some genres withstood the test of time (adult standard & dance pop), you can see how genres begins to diversify by not only introducing new ones, but modifying others as well. 2000 was chosen as the cut-off point because it was the peak of the CD era - which began the transition of music listening through other platforms & mediums.
Shiny Application
Interactive app to generate relationships between variables
Includes two visualization - histogram and scatterplot.
Variables included:
Beats Per Minute (bpm) - The tempo of the song
Energy (nrgy) - The energy of a song - the higher the value, the more energtic song
Danceability (dnce) - The higher the value, the easier it is to dance to this song
Loudness (dB) - The higher the value, the louder the song
Liveness (live) - The higher the value, the more likely the song is a live recording
Valence (val) - The higher the value, the more positive mood for the song
Duration (dur) - The duration of the song
Acousticness (acous) - The higher the value the more acoustic the song is
Speechiness (spch) - The higher the value the more spoken word the song contains
Popularity (pop) - The higher the value the more popular the song is
Variable names and descriptions were created by Spotify - recorded measurements by listeners who reported. You can look and contribute at: Organize Your Music
Map
Global map highlighting countries danceability between years 1942-2019
Music Over The Years - Videos
Conclusions
Through working on this PUG project - useful insights were able to be concluded based upon the datasets utilized and statistical analysis. Quick summary of some interesting findings -
1) If you’re looking to dance, South America is the place to be, in particular: Chile, Argentina, and Bolivia
2) Travel with earplugs if you will be listening to music in India, Australia, or Argentina
3) In the mood for good and positive vibes? Argentina, Columbia, and Australia should be your destination of choice.
Limitations
Between both datasets used, data for earlier years (prior to 1970) was sparse which limited the meaningful conclusion able to be generated between years and time periods. Additionally, some smaller countries did not have any data available and data for those which did was limited at times. To combat this and strengthen analysis - looking for a more holistic dataset or adding on to the current dataset(s) with increased observations from earlier dates and smaller countries would help.