Chapter 1 Introduction

Like everything in life, music has evolved over the years. Over time, the way music is delivered and consumed to its users also saw a significant change - starting from radio to TVs and now via mobile applications and live concerts, we have come a long way! Today, there are very few stores that sell musical CDs, over time people have adapted to this change because of their affection towards music.

This change gave rise to one of the most popular music streaming services in the world: Spotify. According to recent statistics, 35% of subscribers to a music streaming service worldwide have a subscription with Spotify, almost double the proportion who are subscribed to Apple Music, and well above Amazon, Deezer, and other platforms.

Since childhood, Both me and Vedant enjoy listening to songs on a daily basis. We have thousands of songs in our playlist and multiple playlists to choose from depending on the vibe. Hence, visualizing something that we enjoy doing on a regular basis excited us to analyze and visualize the Spotify Dataset.

The data chosen for our project is from the weekly challenge Tidytuesday dataset on Spotify songs. The data comes from Spotify via the spotifyr package authored by Charlie Thompson, Josiah Parry, Donal Phipps, and Tom Wolff. Spotifyr is an R wrapper that was created by pulling track audio features and other information from Spotify’s Web API in bulk. The Dataset entails 23 rows of distinct features and more than 32000 columns of data.

In this project, we are using data visualization for the purpose of gaining insight into the Spotify dataset. For the purpose of helping spotify make music that improves customer satisfaction and also the recognition of less popular artists/bands. The visualizations will also help artists make more informed decisions when choosing the right audio features for their music.