Overview
For our final project we are using data from Spotify to analyze Kanye West and see how he has changed over time through an analysis of his music. We utilized the Spotify API, as well as the spotifyR package to pull data from Spotify on Kanye West. Spotify has provided several predetermined variables based on different aspects of a song. We also pulled Kanye West’s song lyrics from the genius package which will allow us to interpret the sentiment of a particular song or album.
Spotify Variables
Some variables provided by the Spotify API that we will be analyzing include:
valence: A measure from 0.0 to 1.0 describing the musical positiveness conveyed by a track. Tracks with high valence sound more positive (e.g. happy, cheerful, euphoric), while tracks with low valence sound more negative (e.g. sad, depressed, angry).
danceability: how suitable a track is for dancing based on a combination of musical elements including tempo, rhythm stability, beat strength, and overall regularity. A value of 0.0 is least danceable and 1.0 is most danceable.
energy: a measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity. Typically, energetic tracks feel fast, loud, and noisy. For example, death metal has high energy, while a Bach prelude scores low on the scale.
loudness: overall loudness of a track in decibels (dB). Loudness values are averaged across the entire track and are useful for comparing relative loudness of tracks. Loudness is the quality of a sound that is the primary psychological correlate of physical strength (amplitude). Values typical range between -60 and 0 db.
speechiness: the presence of spoken words in a track. The more exclusively speech-like the recording (e.g. talk show, audio book, poetry), the closer to 1.0 the attribute value. Values above 0.66 describe tracks that are probably made entirely of spoken words. Values between 0.33 and 0.66 describe tracks that may contain both music and speech, either in sections or layered, including such cases as rap music. Values below 0.33 most likely represent music and other non-speech-like tracks.
Valence
This plot below corresponds with the valence variable. To reiterate, valence is a measure from 0.0 to 1.0 describing the musical positiveness conveyed by a track. Tracks with high valence sound more positive (e.g. happy, cheerful, euphoric), while tracks with low valence sound more negative (e.g. sad, depressed, angry). The albums are sorted in descending order.
Danceability
The plot below shows which albums are the most danceable. Danceability is how suitable a track is for dancing based on a combination of musical elements including tempo, rhythm stability, beat strength, and overall regularity. A value of 0.0 is least danceable and 1.0 is most danceable. The albums are sorted in descending order, with most danceable album to least danceable album.
Energy
This plot shows the energy within each album, with energy being a measure from 0.0 to 1.0 and represents a perceptual measure of intensity and activity. Typically, energetic tracks feel fast, loud, and noisy. For example, death metal has high energy, while a Bach prelude scores low on the scale. The albums are sorted from those with high energy to those with low energy.
Loudness
The plot below shows the loudness of each album. Loudness is the overall loudness of a track in decibels (dB). Loudness values are averaged across the entire track and are useful for comparing relative loudness of tracks. Loudness is the quality of a sound that is the primary psychological correlate of physical strength (amplitude). Values typical range between -60 and 0 db. The albums are sorted based on the loudest to quieter albums.
Speechiness
The plot below shows the speechiness of each album. Speechiness is defined as the presence of spoken words in a track. The more exclusively speech-like the recording (e.g. talk show, audio book, poetry), the closer to 1.0 the attribute value. Values above 0.66 describe tracks that are probably made entirely of spoken words. Values between 0.33 and 0.66 describe tracks that may contain both music and speech, either in sections or layered, including such cases as rap music. Values below 0.33 most likely represent music and other non-speech-like tracks. Thus, the higher the speechiness value, the more akin to spoken word the album is. Higher speechiness values can be attributed to albums with songs that are more rap-like. The albums are once again sorted in descending order.
Album Change Over Time
These plots below showcase the Spotify variables and how each variable has changed over time. The albums are in chronological order with ‘JESUS IS KING’ being the most recent album and ‘The College Dropout’ being the oldest album.
Valence
Danceability
Energy
Loudness
Speechiness
Most Positive Song
We can see that Kanye’s most positive song according to the valence rating is the song titled “I love Kanye”. This is quite the revelation. Kanye is known to be quite confident in himself and as the plot shows, Kanye does in fact love Kanye.
Most Negative Song
From this plot we can see that Kanye’s most negative song throughout his albums is titled “Violent Crimes”. With a song title like that, we can assume that the valence rating does show that this song is indeed negative.
Most Danceable Song
From this plot below, we can see that Kanye’s most danceable song according to the danceability variable, is titled “All Mine” within the “ye” album.
Most Energetic Song
The most energetic song within Kanye’s discography is called “No More Parties In LA” within the “The Life Of Pablo” album.
Most Speechy Song
The song with the most speechiness is “Skit #3” in the album “Late Registration”. An interesting note is that the top tracks with the most speechiness seem to be titled “Skit” and are differ across three albums.
I’d like to note that these variables may not be entirely “accurate” as all of them were evaluated by an algorithm based on different aspects of each song. As such, computers are not able to grasp the depth of a song as we can as humans. So, feel free to listen to these songs yourself and see whether or not you agree with the plots.
Relationship Between Explicit Songs and Energy Level
This plot illustrates the relationship between explicit lyrics and the energy level of a song. The boxplot shows a higher energy level associated with more explict lyrics.
Relationship Between Tempo and Danceability
This scatter plot of the relationship between danceablity and temp shows little correlation between the two. Each point represents one song and the color corresponds to the album.
Album by Album Text Analysis
Each of Kanye West’s albums are listed below in chronological order starting with the most recent. The worldcloud shows the frequency of words by size, with the most common words being the largest. The second plot separates negative and positve words and displays the most common words in each category.
It is important to note that the sentiment of each word could be different from how it is catagorized in the context of the song. For example, he could be using the words “bad a**" and while both words would be categorized as negative, they are likely positive in context.
Jesus is King - 2019
YE - 2018
Life of Pablo - 2016
Watch The Throne - 2011
Late Registration - 2005
My Beautiful Dark Twisted Fantasy - 2010
Yeezus - 2013
808s and Heartbreak - 2008
Graduation - 2007
The College Dropout - 2004
