Intro
The purpose of this analysis is to elucidate the hidden structure behind the 20 top track of 2022 (deemed by Spotify) by analyzing both the songs’ lyrics and their innert qualities surrounding its composition.
This project is only a descriptive analysis to “sketch” an outline for what a song should possess to have had a chance at success in 2022. This outline is to be compared to in order to predict, it is not predicting in itself.
The project can be extended further by creating a Lyric Generator by training a Long Short Term Memory mode.
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.4.2 ✔ purrr 1.0.1
## ✔ tibble 3.2.1 ✔ dplyr 1.1.2
## ✔ tidyr 1.3.0 ✔ stringr 1.5.0
## ✔ readr 2.1.2 ✔ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(tidytext) #text mining
library(rvest) #web scrape##
## Attaching package: 'rvest'
##
## The following object is masked from 'package:readr':
##
## guess_encodinglibrary(httr) #web scrape
library(wordcloud) #word cloud visualization## Loading required package: RColorBrewerExtracting Lyrics from Genius
The lyrics extracted from Genius coincide to the songs prevalent in Spotify’s most-streamed songs in the U.S. for 2022. The extracted lyrics are then broken down into each section and assigned its own vector. This is to help train the later model in generating lyrics to specific sections of a song
As It Was
hstyles_extracted <- read_html("https://genius.com/Harry-styles-as-it-was-lyrics")
hstyles_lyrics <- html_nodes(hstyles_extracted, ".Dzxov") %>% html_text()
hstyles_lyrics <- paste(hstyles_lyrics[1], hstyles_lyrics[2])
hstyles_intro <- str_extract(hstyles_lyrics, "\\[Intro.*?\\].*?\\[")
hstyles_verse1 <- str_extract(hstyles_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
hstyles_prechorus <- str_extract(hstyles_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
hstyles_chorus <- str_extract(hstyles_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
hstyles_postchorus <- str_extract(hstyles_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
hstyles_verse2 <- str_extract(hstyles_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
hstyles_bridge <- str_extract(hstyles_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
hstyles_outro <- str_extract(hstyles_lyrics, "\\[Outro.*?\\].*")Heat Waves
ganimals_extracted <- read_html("https://genius.com/Glass-animals-heat-waves-lyrics")
ganimals_lyrics <- html_nodes(ganimals_extracted, ".Dzxov") %>% html_text()
ganimals_lyrics <- paste(ganimals_lyrics[1], ganimals_lyrics[2], ganimals_lyrics[3])
ganimals_intro <- str_extract(ganimals_lyrics, "\\[Intro.*?\\].*?\\[")
ganimals_verse1 <- str_extract(ganimals_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
ganimals_prechorus <- str_extract(ganimals_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
ganimals_chorus <- str_extract(ganimals_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
ganimals_postchorus <- str_extract(ganimals_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
ganimals_verse2 <- str_extract(ganimals_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
ganimals_bridge <- str_extract(ganimals_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
ganimals_outro <- str_extract(ganimals_lyrics, "\\[Outro.*?\\].*")Bad Habit
slacy_extracted <- read_html("https://genius.com/Steve-lacy-bad-habit-lyrics")
slacy_lyrics <- html_nodes(slacy_extracted, ".Dzxov") %>% html_text()
slacy_lyrics <- paste(slacy_lyrics[1], slacy_lyrics[2], slacy_lyrics[3])
slacy_intro <- str_extract(slacy_lyrics, "\\[Intro.*?\\].*?\\[")
slacy_verse1 <- str_extract(slacy_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
slacy_prechorus <- str_extract(slacy_lyrics, "\\[Pre-Chorus\\].*?(?=\\[)")
slacy_chorus <- str_extract(slacy_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
slacy_postchorus <- str_extract(slacy_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
slacy_verse2 <- str_extract(slacy_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
slacy_bridge <- str_extract(slacy_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
slacy_instrumental <- str_extract(slacy_lyrics, "\\[Instrumental Break*?\\].*")
slacy_outro <- str_extract(slacy_lyrics, "\\[Outro.*?\\].*")First Class
jharlow_extracted <- read_html("https://genius.com/Jack-harlow-first-class-lyrics")
jharlow_lyrics <- html_nodes(jharlow_extracted, ".Dzxov") %>% html_text()
jharlow_lyrics <- paste(jharlow_lyrics[1], jharlow_lyrics[2])
jharlow_intro <- str_extract(jharlow_lyrics, "\\[Intro.*?\\].*?\\[")
jharlow_verse1 <- str_extract(jharlow_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
jharlow_prechorus <- str_extract(jharlow_lyrics, "\\[Pre-Chorus\\].*?(?=\\[)")
jharlow_chorus <- str_extract(jharlow_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
jharlow_postchorus <- str_extract(jharlow_lyrics, "\\[Post-Chorus\\].*?(?=\\[)")
jharlow_verse2 <- str_extract(jharlow_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
jharlow_bridge <- str_extract(jharlow_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
jharlow_outro <- str_extract(jharlow_lyrics, "\\[Outro.*?\\].*")STAY
klaroi_extracted <- read_html("https://genius.com/The-kid-laroi-and-justin-bieber-stay-lyrics")
klaroi_lyrics <- html_nodes(klaroi_extracted, ".Dzxov") %>% html_text()
klaroi_lyrics <- paste(klaroi_lyrics[1], klaroi_lyrics[2])
klaroi_intro <- str_extract(klaroi_lyrics, "\\[Intro.*?\\].*?\\[")
klaroi_verse1 <- str_extract(klaroi_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
klaroi_prechorus <- str_extract(klaroi_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
klaroi_chorus <- str_extract(klaroi_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
klaroi_postchorus <- str_extract(klaroi_lyrics, "\\[Post-Chorus\\].*?(?=\\[)")
klaroi_verse2 <- str_extract(klaroi_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
klaroi_bridge <- str_extract(klaroi_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
klaroi_outro <- str_extract(klaroi_lyrics, "\\[Outro.*?\\].*")Running Up That Hill (A Deal With God)
kbush_extracted <- read_html("https://genius.com/Kate-bush-running-up-that-hill-a-deal-with-god-lyrics")
kbush_lyrics <- html_nodes(kbush_extracted, ".Dzxov") %>% html_text()
kbush_lyrics <- paste(kbush_lyrics[1], kbush_lyrics[2], kbush_lyrics[3])
kbush_intro <- str_extract(kbush_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
kbush_verse1 <- str_extract(kbush_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
kbush_prechorus <- str_extract(kbush_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
kbush_chorus <- str_extract(kbush_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
kbush_verse2 <- str_extract(kbush_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
kbush_postchorus <- str_extract(kbush_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
kbush_bridge <- str_extract(kbush_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
kbush_outro <- str_extract(kbush_lyrics, "\\[Outro.*?\\].*?(?=\\[)")Sweater Weather
neighbourhood_extracted <- read_html("https://genius.com/The-neighbourhood-sweater-weather-lyrics")
neighbourhood_lyrics <- html_nodes(neighbourhood_extracted, ".Dzxov") %>% html_text()
neighbourhood_lyrics <- paste(neighbourhood_lyrics[1], neighbourhood_lyrics[2], neighbourhood_lyrics[2])
neighbourhood_intro <- str_extract(neighbourhood_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
neighbourhood_verse1 <- str_extract(neighbourhood_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
neighbourhood_prechorus <- str_extract(neighbourhood_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
neighbourhood_chorus <- str_extract(neighbourhood_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
neighbourhood_postchorus <- str_extract(neighbourhood_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
neighbourhood_verse2 <- str_extract(neighbourhood_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
neighbourhood_bridge <- str_extract(neighbourhood_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
neighbourhood_outro <- str_extract(neighbourhood_lyrics, "\\[Outro.*?\\].*?(?=\\[)")Dark Red
slacy2_extracted <- read_html("https://genius.com/Steve-lacy-dark-red-lyrics")
slacy2_lyrics <- html_nodes(slacy2_extracted, ".Dzxov") %>% html_text()
slacy2_lyrics <- paste(slacy2_lyrics[1], slacy2_lyrics[2])
slacy2_intro <- str_extract(slacy2_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
slacy2_verse1 <- str_extract(slacy2_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
slacy2_prechorus <- str_extract(slacy2_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
slacy2_chorus <- str_extract(slacy2_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
slacy2_verse2 <- str_extract(slacy2_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
slacy2_postchorus <- str_extract(slacy2_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
slacy2_bridge <- str_extract(slacy2_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
slacy2_outro <- str_extract(slacy2_lyrics, "\\[Outro.*?\\].*?(?=\\[)")We Don’t Talk About Bruno
encanto_extracted <- read_html("https://genius.com/Carolina-gaitan-mauro-castillo-adassa-rhenzy-feliz-diane-guerrero-and-stephanie-beatriz-we-dont-talk-about-bruno-lyrics")
encanto_lyrics <- html_nodes(encanto_extracted, ".Dzxov") %>% html_text()
encanto_lyrics <- paste(encanto_lyrics[1], encanto_lyrics[2], encanto_lyrics[3])
encanto_intro <- str_extract(encanto_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
encanto_verse1 <- str_extract(encanto_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
encanto_prechorus <- str_extract(encanto_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
encanto_chorus <- str_extract(encanto_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
encanto_verse2 <- str_extract(encanto_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
encanto_postchorus <- str_extract(encanto_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
encanto_bridge <- str_extract(encanto_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
encanto_outro <- str_extract(encanto_lyrics, "\\[Outro.*?\\].*?(?=\\[)")Industry Baby
lilnasx_extracted <- read_html("https://genius.com/Lil-nas-x-and-jack-harlow-industry-baby-lyrics")
lilnasx_lyrics <- html_nodes(lilnasx_extracted, ".Dzxov") %>% html_text()
lilnasx_lyrics <- paste(lilnasx_lyrics[1], lilnasx_lyrics[2], lilnasx_lyrics[3])
lilnasx_intro <- str_extract(lilnasx_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
lilnasx_verse1 <- str_extract(lilnasx_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
lilnasx_prechorus <- str_extract(lilnasx_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
lilnasx_chorus <- str_extract(lilnasx_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
lilnasx_verse2 <- str_extract(lilnasx_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
lilnasx_postchorus <- str_extract(lilnasx_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
lilnasx_bridge <- str_extract(lilnasx_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
lilnasx_outro <- str_extract(lilnasx_lyrics, "\\[Outro.*?\\].*?(?=\\[)")Glimpse of Us
joji_extracted <- read_html("https://genius.com/Joji-glimpse-of-us-lyrics")
joji_lyrics <- html_nodes(joji_extracted, ".Dzxov") %>% html_text()
joji_lyrics <- paste(joji_lyrics[1], joji_lyrics[2], joji_lyrics[3])
joji_intro <- str_extract(joji_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
joji_verse1 <- str_extract(joji_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
joji_prechorus <- str_extract(joji_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
joji_chorus <- str_extract(joji_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
joji_verse2 <- str_extract(joji_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
joji_postchorus <- str_extract(joji_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
joji_bridge <- str_extract(joji_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
joji_outro <- str_extract(joji_lyrics, "\\[Outro.*?\\].*?(?=\\[)")No Role Modelz
jcole_extracted <- read_html("https://genius.com/J-cole-no-role-modelz-lyrics")
jcole_lyrics <- html_nodes(jcole_extracted, ".Dzxov") %>% html_text()
jcole_lyrics <- paste(jcole_lyrics[1], jcole_lyrics[2], jcole_lyrics[3])
jcole_intro <- str_extract(jcole_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
jcole_verse1 <- str_extract(jcole_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
jcole_prechorus <- str_extract(jcole_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
jcole_chorus <- str_extract(jcole_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
jcole_verse2 <- str_extract(jcole_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
jcole_postchorus <- str_extract(jcole_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
jcole_bridge <- str_extract(jcole_lyrics, "\\[Bridge\\].*?(?=\\[)")
jcole_outro <- str_extract(jcole_lyrics, "\\[Outro.*?\\].*?(?=\\[)")Super Gremlin
kodak_extracted <- read_html("https://genius.com/Kodak-black-super-gremlin-lyrics")
kodak_lyrics <- html_nodes(kodak_extracted, ".Dzxov") %>% html_text()
kodak_lyrics <- paste(kodak_lyrics[1], kodak_lyrics[2], kodak_lyrics[3])
kodak_intro <- str_extract(kodak_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
kodak_verse1 <- str_extract(kodak_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
kodak_prechorus <- str_extract(kodak_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
kodak_chorus <- str_extract(kodak_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
kodak_verse2 <- str_extract(kodak_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
kodak_postchorus <- str_extract(kodak_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
kodak_bridge <- str_extract(kodak_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
kodak_outro <- str_extract(kodak_lyrics, "\\[Outro.*?\\].*?(?=\\[)")Knife Talk
drake_extracted <- read_html("https://genius.com/Drake-knife-talk-lyrics")
drake_lyrics <- html_nodes(drake_extracted, ".Dzxov") %>% html_text()
drake_lyrics <- paste(drake_lyrics[1], drake_lyrics[2], drake_lyrics[3])
drake_intro <- str_extract(drake_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
drake_verse1 <- str_extract(drake_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
drake_prechorus <- str_extract(drake_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
drake_chorus <- str_extract(drake_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
drake_verse2 <- str_extract(drake_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
drake_postchorus <- str_extract(drake_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
drake_bridge <- str_extract(drake_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
drake_outro <- str_extract(drake_lyrics, "\\[Outro.*?\\].*?(?=\\[)")WAIT FOR U
future_extracted <- read_html("https://genius.com/Future-wait-for-u-lyrics")
future_lyrics <- html_nodes(future_extracted, ".Dzxov") %>% html_text()
future_lyrics <- paste(future_lyrics[1], future_lyrics[2], future_lyrics[3])
future_intro <- str_extract(future_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
future_verse1 <- str_extract(future_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
future_prechorus <- str_extract(future_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
future_chorus <- str_extract(future_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
future_verse2 <- str_extract(future_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
future_postchorus <- str_extract(future_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
future_bridge <- str_extract(future_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
future_outro <- str_extract(future_lyrics, "\\[Outro.*?\\].*?(?=\\[)")About Damn Time
lizzo_extracted <- read_html("https://genius.com/Lizzo-about-damn-time-lyrics")
lizzo_lyrics <- html_nodes(lizzo_extracted, ".Dzxov") %>% html_text()
lizzo_lyrics <- paste(lizzo_lyrics[1], lizzo_lyrics[2], lizzo_lyrics[3])
lizzo_intro <- str_extract(lizzo_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
lizzo_verse1 <- str_extract(lizzo_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
lizzo_prechorus <- str_extract(lizzo_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
lizzo_chorus <- str_extract(lizzo_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
lizzo_verse2 <- str_extract(lizzo_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
lizzo_postchorus <- str_extract(lizzo_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
lizzo_bridge <- str_extract(lizzo_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
lizzo_outro <- str_extract(lizzo_lyrics, "\\[Outro.*?\\].*?(?=\\[)")Enemy
idragons_extracted <- read_html("https://genius.com/Imagine-dragons-and-jid-enemy-lyrics")
idragons_lyrics <- html_nodes(idragons_extracted, ".Dzxov") %>% html_text()
idragons_lyrics <- paste(idragons_lyrics[1], idragons_lyrics[2], idragons_lyrics[3])
idragons_intro <- str_extract(idragons_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
idragons_verse1 <- str_extract(idragons_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
idragons_prechorus <- str_extract(idragons_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
idragons_chorus <- str_extract(idragons_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
idragons_verse2 <- str_extract(idragons_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
idragons_postchorus <- str_extract(idragons_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
idragons_bridge <- str_extract(idragons_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
idragons_outro <- str_extract(idragons_lyrics, "\\[Outro\\].*?(?=\\[)")505
amonkeys_extracted <- read_html("https://genius.com/Arctic-monkeys-505-lyrics")
amonkeys_lyrics <- html_nodes(amonkeys_extracted, ".Dzxov") %>% html_text()
amonkeys_lyrics <- paste(amonkeys_lyrics[1], amonkeys_lyrics[2], amonkeys_lyrics[3])
amonkeys_intro <- str_extract(amonkeys_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
amonkeys_verse1 <- str_extract(amonkeys_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
amonkeys_prechorus <- str_extract(amonkeys_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
amonkeys_chorus <- str_extract(amonkeys_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
amonkeys_verse2 <- str_extract(amonkeys_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
amonkeys_postchorus <- str_extract(amonkeys_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
amonkeys_bridge <- str_extract(amonkeys_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
amonkeys_outro <- str_extract(amonkeys_lyrics, "\\[Outro.*?\\].*?(?=\\[)")I Love You So
walters_extracted <- read_html("https://genius.com/The-walters-i-love-you-so-lyrics")
walters_lyrics <- html_nodes(walters_extracted, ".Dzxov") %>% html_text()
walters_lyrics <- paste(walters_lyrics[1], walters_lyrics[2], walters_lyrics[3])
walters_intro <- str_extract(walters_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
walters_verse1 <- str_extract(walters_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
walters_prechorus <- str_extract(walters_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
walters_chorus <- str_extract(walters_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
walters_verse2 <- str_extract(walters_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
walters_postchorus <- str_extract(walters_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
walters_bridge <- str_extract(walters_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
walters_outro <- str_extract(walters_lyrics, "\\[Outro.*?\\].*?(?=\\[)")good 4 u
orodrigo_extracted <- read_html("https://genius.com/Olivia-rodrigo-good-4-u-lyrics")
orodrigo_lyrics <- html_nodes(orodrigo_extracted, ".Dzxov") %>% html_text()
orodrigo_lyrics <- paste(orodrigo_lyrics[1], orodrigo_lyrics[2], orodrigo_lyrics[3])
orodrigo_intro <- str_extract(orodrigo_lyrics, "\\[Intro.*?\\].*?(?=\\[)")
orodrigo_verse1 <- str_extract(orodrigo_lyrics, "\\[Verse 1.*?\\].*?(?=\\[)")
orodrigo_prechorus <- str_extract(orodrigo_lyrics, "\\[Pre-Chorus.*?\\].*?(?=\\[)")
orodrigo_chorus <- str_extract(orodrigo_lyrics, "\\[Chorus.*?\\].*?(?=\\[)")
orodrigo_verse2 <- str_extract(orodrigo_lyrics, "\\[Verse 2.*?\\].*?(?=\\[)")
orodrigo_postchorus <- str_extract(orodrigo_lyrics, "\\[Post-Chorus.*?\\].*?(?=\\[)")
orodrigo_bridge <- str_extract(orodrigo_lyrics, "\\[Bridge.*?\\].*?(?=\\[)")
orodrigo_outro <- str_extract(orodrigo_lyrics, "\\[Outro.*?\\].*?(?=\\[)")combining lyrics by sections
intro <- c(hstyles_intro, ganimals_intro, slacy_intro, jharlow_intro, klaroi_intro, kbush_intro, neighbourhood_intro, slacy2_intro, encanto_intro, lilnasx_intro, joji_intro, jcole_intro, kodak_intro, drake_intro, future_intro, lizzo_intro, idragons_intro, amonkeys_intro, walters_intro, orodrigo_intro)
intro <- str_remove_all(intro, "\\[Intro.*?]|,|\\[,|\\(|\\)")
intro <- gsub("([A-Z])", "\n\\1", intro)
verse1 <- c(hstyles_verse1, ganimals_verse1, slacy_verse1, jharlow_verse1, klaroi_verse1, kbush_verse1, neighbourhood_verse1, slacy2_verse1, encanto_verse1, lilnasx_verse1, joji_verse1, jcole_verse1, kodak_verse1, drake_verse1, future_verse1, lizzo_verse1, idragons_verse1, amonkeys_verse1, walters_verse1, orodrigo_verse1)
verse1 <- str_remove_all(verse1, "\\[Verse 1.*?]|,|\\[,|\\(|\\)")
verse1 <- gsub("([A-Z])", " \\1", verse1)
pre_chorus <- c(hstyles_prechorus, ganimals_prechorus, slacy_prechorus, jharlow_prechorus, klaroi_prechorus, kbush_prechorus, neighbourhood_prechorus, slacy2_prechorus, encanto_prechorus, lilnasx_prechorus, joji_prechorus, jcole_prechorus, kodak_prechorus, drake_prechorus, future_prechorus, lizzo_prechorus, idragons_prechorus, amonkeys_prechorus, walters_prechorus, orodrigo_prechorus)
pre_chorus <- str_remove_all(pre_chorus, "\\[Pre-Chorus.*?]|,|\\[,|\\(|\\)")
pre_chorus <- gsub("([A-Z])", "\n\\1", pre_chorus)
chorus <- c(hstyles_chorus, ganimals_chorus, slacy_chorus, jharlow_chorus, klaroi_chorus, kbush_chorus, neighbourhood_chorus, slacy2_chorus, encanto_chorus, lilnasx_chorus, joji_chorus, jcole_chorus, kodak_chorus, drake_chorus, future_chorus, lizzo_chorus, idragons_chorus, amonkeys_chorus, walters_chorus, orodrigo_chorus)
chorus <- str_remove_all(chorus, "\\[Chorus.*?]|[^[:alnum:]\\s]")
chorus <- gsub("([A-Z])", "\n\\1", chorus)
post_chorus <- c(hstyles_postchorus, ganimals_postchorus, slacy_postchorus, jharlow_postchorus, klaroi_postchorus, kbush_postchorus, neighbourhood_postchorus, slacy2_postchorus, encanto_postchorus, lilnasx_postchorus, joji_postchorus, jcole_postchorus, kodak_postchorus, drake_postchorus, future_postchorus, lizzo_postchorus, idragons_postchorus, amonkeys_postchorus, walters_postchorus, orodrigo_postchorus)
post_chorus <- str_remove_all(post_chorus, "\\[Post-Chorus.*?]|[^[:alnum:]\\s]")
post_chorus <- gsub("([A-Z])", "\n\\1", post_chorus)
verse2 <- c(hstyles_verse2, ganimals_verse2, slacy_verse2, jharlow_verse2, klaroi_verse2, kbush_verse2, neighbourhood_verse2, slacy2_verse2, encanto_verse2, lilnasx_verse2, joji_verse2, jcole_verse2, kodak_verse2, drake_verse2, future_verse2, lizzo_verse2, idragons_verse2, amonkeys_verse2, walters_verse2, orodrigo_verse2)
verse2 <- str_remove_all(verse2, "\\[Verse 2.*?]|\\(|\\)|\"|,|\\[,|\\(|\\)")
verse2 <- gsub("([A-Z])", " \\1", verse2)
bridge <- c(hstyles_bridge, ganimals_bridge, slacy_bridge, jharlow_bridge, klaroi_bridge, kbush_bridge, neighbourhood_bridge, slacy2_bridge, encanto_bridge, lilnasx_bridge, joji_bridge, jcole_bridge, kodak_bridge, drake_bridge, future_bridge, lizzo_bridge, idragons_bridge, amonkeys_bridge, walters_bridge, orodrigo_bridge)
bridge <- str_remove_all(bridge, "\\[Bridge.*?]|,|\\[,|\\(|\\)")
bridge <- gsub("([A-Z])", "\n\\1", bridge)
outro <- c(hstyles_outro, ganimals_outro, slacy_outro, jharlow_outro, klaroi_outro, kbush_outro, neighbourhood_outro, slacy2_outro, encanto_outro, lilnasx_outro, joji_outro, jcole_outro, kodak_outro, drake_outro, future_outro, lizzo_outro, idragons_outro, amonkeys_outro, walters_outro, orodrigo_outro)
outro <- str_remove_all(outro, "\\[Outro.*?]|,|\\[,|\\(|\\)")
outro <- gsub("([A-Z])", "\n\\1", outro)data <- data.frame(
Artist = c("Harry Styles", "Glass Animals", "Steve Lacy", "Jack Harlow", "The Kid Laroi", "Kate Bush", "the Neighbourhood", "Steve Lacy", "Carolina Gaitan", "Lil Nas X", "Joji", "J.Cole", "Kodak Black", "Drake", "Future", "Lizzo", "Imagine Dragons", "Arctic Monkeys", "The Walters", "Olivia Rodrigo"),
Song = c("As It Was", "Heat Waves", "Bad Habit", "First Class", "STAY (feat. Justin Bieber)", "Running Up That Hill (A Deal With God)", "Sweater Weather", "Dark Red", "We Don’t Talk About Bruno", "Industry Baby (feat. Jack Harlow)", "Glimpse of Us", "No Role Modelz", "Super Gremlin", "Knife Talk (with 21 Savage ft. Project Pat", "WAIT FOR U (feat. Drake & Tems)", "About Damn Time", "Enemy (with JID)", "505", "I Love You So", "good 4 u" )
)
data$intro <- intro
data$verse1 <- verse1
data$prechorus <- pre_chorus
data$chorus <- chorus
data$postchorus <- post_chorus
data$verse2 <- verse2
data$bridge <- bridge
data$outro <- outroWide -> Long DataFrame
pivot_data <- data %>%
pivot_longer(cols = -c(Artist, Song), names_to = "Section", values_to = "Lyrics") %>%
mutate(Lyrics = str_split(Lyrics, pattern = "\\s+")) %>%
unnest()## Warning: `cols` is now required when using `unnest()`.
## ℹ Please use `cols = c(Lyrics)`.pivot_data <- pivot_data %>%
filter(Lyrics != "")
pivot_data## # A tibble: 5,036 × 4
## Artist Song Section Lyrics
## <chr> <chr> <chr> <chr>
## 1 Harry Styles As It Was intro Come
## 2 Harry Styles As It Was intro on
## 3 Harry Styles As It Was intro Harry
## 4 Harry Styles As It Was intro we
## 5 Harry Styles As It Was intro wanna
## 6 Harry Styles As It Was intro say
## 7 Harry Styles As It Was intro goodnight
## 8 Harry Styles As It Was intro to
## 9 Harry Styles As It Was intro you[
## 10 Harry Styles As It Was verse1 Holdin'
## # ℹ 5,026 more rowsremoving the stop words
pivot_data_cleaned <- pivot_data # pivot_data_cleaned is a duplicated pivot_data with the removed stop words
stopwords <- tolower(stop_words$word)
pivot_data_cleaned$Lyrics <- ifelse(tolower(pivot_data$Lyrics) %in% stopwords, "", pivot_data$Lyrics)
pivot_data_cleaned <- pivot_data_cleaned %>%
filter(Lyrics != "")
more_stopwords <- c("yo", "yeah", "i'ma", "ooh-woah", "l.", "a.", "youre", "im", "id", "dont", "alright", "woah", "ooh", "oh-oh-oh", "a—[", "ah", "ayy", "d-", "baby", "uh", "hah", "nigga", "niggas", "damn", "shit", "y'all", "ch", "'bout", "gettin'", "comin'", "babe", "mmm")
pivot_data_cleaned <- pivot_data_cleaned %>%
filter(!Lyrics %in% more_stopwords)Intro
library(knitr)
library(kableExtra)##
## Attaching package: 'kableExtra'## The following object is masked from 'package:dplyr':
##
## group_rowspivot_data_cleaned <- pivot_data_cleaned %>%
filter(!Lyrics %in% more_stopwords)
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "intro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 30.2avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "intro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 13.7avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "intro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 10.8dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 30.18182 |
| Average Excluding Stopwords | 13.66667 |
| Average Unique Words (excluding stopwords) | 10.77778 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Intro Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(5, 0, 5, 0)),
plot.title.position = "plot")
pivot_data_intro <- pivot_data_cleaned %>%
filter(Section == "intro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_intro## # A tibble: 92 × 2
## Lyrics n
## <chr> <int>
## 1 gang 4
## 2 heat 4
## 3 jacob 4
## 4 wait 4
## 5 couple 3
## 6 time 3
## 7 yeah 3
## 8 ayy 2
## 9 d- 2
## 10 dummies 2
## # ℹ 82 more rowspivot_data_intro <- pivot_data_intro %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_intro %>%
filter(n > 1) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity", fill = "#22BBFE") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Intros") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_intro %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("bing"), by = "word") ## # A tibble: 9 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 shimmer 2 negative
## 2 bleed 1 negative
## 3 cheat 1 negative
## 4 error 1 negative
## 5 fell 1 negative
## 6 lose 1 negative
## 7 pretty 1 positive
## 8 quicker 1 positive
## 9 tops 1 positiveVerse 1
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "verse1") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 65.4avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "verse1") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 20.4avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "verse1") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 18.3dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 65.36842 |
| Average Excluding Stopwords | 20.42105 |
| Average Unique Words (excluding stopwords) | 18.26316 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Verse1 Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(10, 0, 10, 0)),
plot.title.position = "plot")
pivot_data_verse1 <- pivot_data_cleaned %>%
filter(Section == "verse1") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_verse1 <- pivot_data_verse1 %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_verse1 %>%
filter(n > 2) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity", fill = "#22BBFE") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Verse 1") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_verse1 %>%
filter(n > 2)## # A tibble: 12 × 2
## Lyrics n
## <chr> <int>
## 1 gang 8
## 2 feel 6
## 3 day 4
## 4 girl 4
## 5 time 4
## 6 wanna 4
## 7 world 4
## 8 guess 3
## 9 leave 3
## 10 move 3
## 11 spinnin' 3
## 12 sweet 3pivot_data_verse1 %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("bing"), by = "word") ## # A tibble: 34 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 sweet 3 positive
## 2 bad 2 negative
## 3 bitch 2 negative
## 4 fuck 2 negative
## 5 hard 2 negative
## 6 helped 2 positive
## 7 hurt 2 negative
## 8 perfect 2 positive
## 9 wasted 2 negative
## 10 adore 1 positive
## # ℹ 24 more rowsPre-Chorus
Across the pre-chorus of the sampled songs, there’s an average of 23 words with stop words included in a pre-chorus, 8 words with stop words removed where 6 of those words are unique to the verse.
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "prechorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 22.4avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "prechorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 4.29avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "prechorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 3.57dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 22.375000 |
| Average Excluding Stopwords | 4.285714 |
| Average Unique Words (excluding stopwords) | 3.571429 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(10, 0, 10, 0)),
plot.title.position = "plot")
pivot_data_prechorus <- pivot_data_cleaned %>%
filter(Section == "prechorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_intro## # A tibble: 87 × 2
## Lyrics n
## <chr> <int>
## 1 gang 4
## 2 heat 4
## 3 jacob 4
## 4 wait 4
## 5 couple 3
## 6 time 3
## 7 dummies 2
## 8 night 2
## 9 road 2
## 10 shimmer 2
## # ℹ 77 more rowspivot_data_prechorus <- pivot_data_prechorus %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_prechorus %>%
filter(n > 1) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity", fill = "#22BBFE") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Pre-Chorus") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_prechorus %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("bing"), by = "word") ## # A tibble: 7 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 love 2 positive
## 2 bitch 1 negative
## 3 fine 1 positive
## 4 funny 1 negative
## 5 hate 1 negative
## 6 lame 1 negative
## 7 lost 1 negativeChorus
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "chorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 53.6avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "chorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 14.3avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "chorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 9.26dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 53.631579 |
| Average Excluding Stopwords | 14.263158 |
| Average Unique Words (excluding stopwords) | 9.263158 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Chorus Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(10, 0, 10, 0)),
plot.title.position = "plot")
pivot_data_chorus <- pivot_data_cleaned %>%
filter(Section == "chorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_chorus## # A tibble: 153 × 2
## Lyrics n
## <chr> <int>
## 1 im 18
## 2 gang 8
## 3 wait 7
## 4 dont 6
## 5 wanna 6
## 6 time 5
## 7 class 4
## 8 love 4
## 9 save 4
## 10 saved 4
## # ℹ 143 more rowspivot_data_chorus <- pivot_data_chorus %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_chorus %>%
filter(n > 2) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity", fill = "#22BBFE") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Chorus") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_chorus %>%
filter(n > 2) %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("nrc"), by = "word") ## # A tibble: 11 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 gang 8 anger
## 2 gang 8 fear
## 3 gang 8 negative
## 4 wait 7 anticipation
## 5 wait 7 negative
## 6 time 5 anticipation
## 7 love 4 joy
## 8 love 4 positive
## 9 save 4 joy
## 10 save 4 positive
## 11 save 4 trustPost-Chorus
Across the post-chorus of the sampled songs, there’s an average of 40 words with stop words included in a pre-chorus, 10 words with stop words removed where 8 of those words are unique to the verse.
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "postchorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 39.7avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "postchorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 8.33avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "postchorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 7.33dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 39.666667 |
| Average Excluding Stopwords | 8.333333 |
| Average Unique Words (excluding stopwords) | 7.333333 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(10, 0, 10, 0)),
plot.title.position = "plot")
pivot_data_postchorus <- pivot_data_cleaned %>%
filter(Section == "postchorus") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_postchorus## # A tibble: 21 × 2
## Lyrics n
## <chr> <int>
## 1 im 3
## 2 yeah 3
## 3 ate 1
## 4 bet 1
## 5 bitch 1
## 6 blick 1
## 7 caught 1
## 8 fake 1
## 9 gremlin 1
## 10 kit 1
## # ℹ 11 more rowspivot_data_postchorus <- pivot_data_postchorus %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_postchorus %>%
filter(n > 1) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Post-Chorus") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_postchorus %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("bing"), by = "word") ## # A tibble: 3 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 bitch 1 negative
## 2 fake 1 negative
## 3 unhappy 1 negativeVerse 2
On average, the second verse of the sampled songs contain 73.5 words with stop words and 25 words without where 22 of those words are unique
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "verse2") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 86.3avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "verse2") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 27.9avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "verse2") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 25.6dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 86.31579 |
| Average Excluding Stopwords | 27.89474 |
| Average Unique Words (excluding stopwords) | 25.57895 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Verse 2 Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(10, 0, 10, 0)),
plot.title.position = "plot")
pivot_data_verse2 <- pivot_data_cleaned %>%
filter(Section == "verse2") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_verse2 <- pivot_data_verse2 %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_verse2 %>%
filter(n > 2) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity", fill = "#22BBFE") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Verse 2") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_verse2 %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("nrc"), by = "word") ## # A tibble: 208 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 wait 9 anticipation
## 2 wait 9 negative
## 3 time 7 anticipation
## 4 guess 4 surprise
## 5 leave 3 negative
## 6 leave 3 sadness
## 7 leave 3 surprise
## 8 love 3 joy
## 9 love 3 positive
## 10 start 3 anticipation
## # ℹ 198 more rowsBridge
Across the intros of the sampled songs, there’s an average of 53 words in an intro with stop words included, 23 words with stop words removed where 12 of those words are unique to the verse.
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "bridge") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 50.1avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "bridge") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 14.5avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "bridge") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 9.25dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 50.125 |
| Average Excluding Stopwords | 14.500 |
| Average Unique Words (excluding stopwords) | 9.250 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(10, 0, 10, 0)),
plot.title.position = "plot")
pivot_data_bridge <- pivot_data_cleaned %>%
filter(Section == "bridge") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_bridge## # A tibble: 74 × 2
## Lyrics n
## <chr> <int>
## 1 tonight 14
## 2 woah 5
## 3 emotional 4
## 4 time 3
## 5 bad 2
## 6 cared 2
## 7 comin' 2
## 8 darlin' 2
## 9 decide 2
## 10 fakin' 2
## # ℹ 64 more rowspivot_data_bridge <- pivot_data_bridge %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_bridge %>%
filter(n > 2) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity", fill = "#22BBFE") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Bridge") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_bridge %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("bing"), by = "word") ## # A tibble: 14 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 bad 2 negative
## 2 woo 2 positive
## 3 wound 2 negative
## 4 wrong 2 negative
## 5 angel 1 positive
## 6 apathy 1 negative
## 7 comfortable 1 positive
## 8 fine 1 positive
## 9 fuck 1 negative
## 10 hard 1 negative
## 11 perfectly 1 positive
## 12 smile 1 positive
## 13 steal 1 negative
## 14 wow 1 positiveOutro
Across the outros of the sampled songs, there’s an average of 35 words in an intro with stop words included, 14 words with stop words removed where 9 of those words are unique to the verse.
avg_words_with_stopwords <- pivot_data %>%
filter(Section == "outro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words_with_stopwords = mean(average_words))
avg_words_with_stopwords## # A tibble: 1 × 1
## mean_average_words_with_stopwords
## <dbl>
## 1 34.7avg_words_no_stopwords <- pivot_data_cleaned %>%
filter(Section == "outro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_words = mean(length(Lyrics))) %>%
summarise(mean_average_words = mean(average_words))
avg_words_no_stopwords## # A tibble: 1 × 1
## mean_average_words
## <dbl>
## 1 12avg_unique_words_verse <- pivot_data_cleaned %>%
filter(Section == "outro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Artist) %>%
summarise(average_unique_words = mean(length(unique(Lyrics)))) %>%
summarise(mean_average_unique_words = mean(average_unique_words))
avg_unique_words_verse## # A tibble: 1 × 1
## mean_average_unique_words
## <dbl>
## 1 8dtaf <- c(avg_words_with_stopwords, avg_words_no_stopwords, avg_unique_words_verse)
dtaf <- data.frame(dtaf)
dtaf <- pivot_longer(dtaf, everything(), names_to = "Average", values_to = "Value")
dtaf$Average <- c("Average Including Stopwords", "Average Excluding Stopwords", "Average Unique Words (excluding stopwords)")
kable(dtaf, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Average | Value |
|---|---|
| Average Including Stopwords | 34.66667 |
| Average Excluding Stopwords | 12.00000 |
| Average Unique Words (excluding stopwords) | 8.00000 |
df <- data.frame(
Component = c("Average Words (+ SW)",
"Average Words (-SW)",
"Average Unique Words (-SW)"),
Count = c(avg_words_with_stopwords$mean_average_words_with_stopwords, avg_words_no_stopwords$mean_average_words, avg_unique_words_verse$mean_average_unique_words)
)
ggplot(df, aes(x = Component, y = Count, fill = Component)) +
geom_bar(stat = "identity") +
labs(title = "Outro Word Composition", y = "Count") +
theme_minimal() +
theme(plot.margin = margin(5, 5, 5, 5, "mm"),
legend.position = "none",
plot.title = element_text(hjust = 0.5, vjust = 0.5, size = 14, margin = margin(10, 0, 10, 0)),
plot.title.position = "plot")
pivot_data_outro <- pivot_data_cleaned %>%
filter(Section == "outro") %>%
mutate(Lyrics = tolower(Lyrics)) %>%
group_by(Lyrics) %>%
summarise(n = n()) %>%
arrange(desc(n))
pivot_data_outro## # A tibble: 24 × 2
## Lyrics n
## <chr> <int>
## 1 heat 4
## 2 biscuits 2
## 3 gravy 2
## 4 mirror 2
## 5 road 2
## 6 shimmer 2
## 7 swimmin' 2
## 8 vision 2
## 9 waves 2
## 10 wigglin' 2
## # ℹ 14 more rowspivot_data_bridge <- pivot_data_bridge %>%
filter(!Lyrics %in% more_stopwords)
pivot_data_outro %>%
filter(n > 1) %>%
ggplot(aes(Lyrics, n)) +
geom_bar(stat = "identity", fill = "#22BBFE") +
xlab("Lyrics") +
ylab("Frequency") +
ggtitle("Words Frequently Used in Outro") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5)) +
coord_flip()
pivot_data_outro %>%
rename(word = Lyrics) %>%
inner_join(get_sentiments("bing"), by = "word") ## # A tibble: 7 × 3
## word n sentiment
## <chr> <int> <chr>
## 1 shimmer 2 negative
## 2 beg 1 negative
## 3 crazy 1 negative
## 4 fuck 1 negative
## 5 lose 1 negative
## 6 miss 1 negative
## 7 stupid 1 negativeCompositional Qualities to the Sampled Music
#Extracting Song Qualities from Spotify API Using Spotify API, the qualitative factors of each song is extracted (danceability, energy, tempo, key, etc)
library(spotifyr)##
## Attaching package: 'spotifyr'## The following object is masked from 'package:tidytext':
##
## tidylibrary(knitr)
Sys.setenv(SPOTIFY_CLIENT_ID = "8a9ee4f456f244fba50f78ca701c4bdd")
Sys.setenv(SPOTIFY_CLIENT_SECRET = "d7aab5410fa342859cb7a30953003e06")#Harry Styles' "As It was"
hstyles_qual <- get_track_audio_features("4LRPiXqCikLlN15c3yImP7")
hstyles_qual$Artist <- "Harry Styles"
hstyles_qual$Song <- "As It Was"
hstyles_qual <- hstyles_qual[, c(ncol(hstyles_qual) - 1, ncol(hstyles_qual), 1:(ncol(hstyles_qual) - 2))]#Glass Animals "Heat Waves"
ganimals_qual <- get_track_audio_features("3USxtqRwSYz57Ewm6wWRMp")
ganimals_qual$Artist <- "Glass Animals"
ganimals_qual$Song <- "Heat Waves"
ganimals_qual <- ganimals_qual[, c(ncol(ganimals_qual) - 1, ncol(ganimals_qual), 1:(ncol(ganimals_qual) - 2))]#Steve Lacy "Bad Habit"
slacy_qual <- get_track_audio_features("3EaJDYHA0KnX88JvDhL9oa")
slacy_qual$Artist <- "Steve Lacy"
slacy_qual$Song <- "Bad Habit"
slacy_qual <- slacy_qual[, c(ncol(slacy_qual) - 1, ncol(slacy_qual), 1:(ncol(slacy_qual) - 2))]#Jack Harlow "First Class"
jharlow_qual <- get_track_audio_features("0wHFktze2PHC5jDt3B17DC")
jharlow_qual$Artist <- "Jack Harlow"
jharlow_qual$Song <- "First Class"
jharlow_qual <- jharlow_qual[, c(ncol(jharlow_qual) - 1, ncol(jharlow_qual), 1:(ncol(jharlow_qual) - 2))]
jharlow_qual## # A tibble: 1 × 20
## Artist Song danceability energy key loudness mode speechiness acousticness
## <chr> <chr> <dbl> <dbl> <int> <dbl> <int> <dbl> <dbl>
## 1 Jack … Firs… 0.902 0.582 5 -5.90 0 0.109 0.111
## # ℹ 11 more variables: instrumentalness <dbl>, liveness <dbl>, valence <dbl>,
## # tempo <dbl>, type <chr>, id <chr>, uri <chr>, track_href <chr>,
## # analysis_url <chr>, duration_ms <int>, time_signature <int>#Kid Laroi "STAY" (feat Justin Beiber)
klaroi_qual <- get_track_audio_features("5HCyWlXZPP0y6Gqq8TgA20")
klaroi_qual$Artist <- "The Kid Laroi"
klaroi_qual$Song <- "STAY (feat. Justin Bieber)"
klaroi_qual <- klaroi_qual[, c(ncol(klaroi_qual) - 1, ncol(klaroi_qual), 1:(ncol(klaroi_qual) - 2))]kbush_qual <- get_track_audio_features("1PtQJZVZIdWIYdARpZRDFO")
kbush_qual$Artist <- "Kate Bush"
kbush_qual$Song <- "Running Up That Hill (A Deal With God)"
kbush_qual <- kbush_qual[, c(ncol(kbush_qual) - 1, ncol(kbush_qual), 1:(ncol(kbush_qual) - 2))]neighbourhood_qual <- get_track_audio_features("2QjOHCTQ1Jl3zawyYOpxh6")
neighbourhood_qual$Artist <- "the Neighbourhood"
neighbourhood_qual$Song <- "Sweater Weather"
neighbourhood_qual <- neighbourhood_qual[, c(ncol(neighbourhood_qual) - 1, ncol(neighbourhood_qual), 1:(ncol(neighbourhood_qual) - 2))]slacy2_qual <- get_track_audio_features("3EaJDYHA0KnX88JvDhL9oa")
slacy2_qual$Artist <- "Steve Lacy"
slacy2_qual$Song <- "Dark Red"
slacy2_qual <- slacy2_qual[, c(ncol(slacy2_qual) - 1, ncol(slacy2_qual), 1:(ncol(slacy2_qual) - 2))]encanto_qual <- get_track_audio_features("2xJxFP6TqMuO4Yt0eOkMz")
encanto_qual$Artist <- "Carolina Gaitan"
encanto_qual$Song <- "We Don’t Talk About Bruno"
encanto_qual <- encanto_qual[, c(ncol(encanto_qual) - 1, ncol(encanto_qual), 1:(ncol(encanto_qual) - 2))]lilnasx_qual <- get_track_audio_features("5Z9KJZvQzH6PFmb8SNkxuk")
lilnasx_qual$Artist <- "Lil Nas X"
lilnasx_qual$Song <- "Industry Baby (feat. Jack Harlow)"
lilnasx_qual <- lilnasx_qual[, c(ncol(lilnasx_qual) - 1, ncol(lilnasx_qual), 1:(ncol(lilnasx_qual) - 2))]joji_qual <- get_track_audio_features("4ewazQLXFTDC8XvCbhvtXs")
joji_qual$Artist <- "Joji"
joji_qual$Song <- "Glimpse of Us"
joji_qual <- joji_qual[, c(ncol(joji_qual) - 1, ncol(joji_qual), 1:(ncol(joji_qual) - 2))]jcole_qual <- get_track_audio_features("68Dni7IE4VyPkTOH9mRWHr")
jcole_qual$Artist <- "J.Cole"
jcole_qual$Song <- "No Role Modelz"
jcole_qual <- jcole_qual[, c(ncol(jcole_qual) - 1, ncol(jcole_qual), 1:(ncol(jcole_qual) - 2))]kodak_qual <- get_track_audio_features("1Y5Jvi3eLi4Chwqch9GMem")
kodak_qual$Artist <- "Kodak Black"
kodak_qual$Song <- "Super Gremlin"
kodak_qual <- kodak_qual[, c(ncol(kodak_qual) - 1, ncol(kodak_qual), 1:(ncol(kodak_qual) - 2))]
kodak_qual## # A tibble: 1 × 20
## Artist Song danceability energy key loudness mode speechiness acousticness
## <chr> <chr> <dbl> <dbl> <int> <dbl> <int> <dbl> <dbl>
## 1 Kodak… Supe… 0.825 0.414 2 -6.63 1 0.144 0.00265
## # ℹ 11 more variables: instrumentalness <int>, liveness <dbl>, valence <dbl>,
## # tempo <dbl>, type <chr>, id <chr>, uri <chr>, track_href <chr>,
## # analysis_url <chr>, duration_ms <int>, time_signature <int>drake_qual <- get_track_audio_features("2BcMwX1MPV6ZHP4tUT9uq6")
drake_qual$Artist <- "Drake"
drake_qual$Song <- "Knife Talk (with 21 Savage ft. Project Pat)"
drake_qual <- drake_qual[, c(ncol(drake_qual) - 1, ncol(drake_qual), 1:(ncol(drake_qual) - 2))]future_qual <- get_track_audio_features("59nOXPmaKlBfGMDeOVGrIK")
future_qual$Artist <- "Future"
future_qual$Song <- "WAIT FOR U (feat. Drake & Tems)"
future_qual <- future_qual[, c(ncol(future_qual) - 1, ncol(future_qual), 1:(ncol(future_qual) - 2))]lizzo_qual <- get_track_audio_features("6HMtHNpW6YPi1hrw9tgF8P")
lizzo_qual$Artist <- "Lizzo"
lizzo_qual$Song <- "About Damn Time"
lizzo_qual <- lizzo_qual[, c(ncol(lizzo_qual) - 1, ncol(lizzo_qual), 1:(ncol(lizzo_qual) - 2))]idragons_qual <- get_track_audio_features("3CIyK1V4JEJkg02E4EJnDl")
idragons_qual$Artist <- "Imagine Dragons"
idragons_qual$Song <- "Enemy (with JID) "
idragons_qual <- idragons_qual[, c(ncol(idragons_qual) - 1, ncol(idragons_qual), 1:(ncol(idragons_qual) - 2))]amonkeys_qual <- get_track_audio_features("58ge6dfP91o9oXMzq3XkIS")
amonkeys_qual$Artist <- "Arctic Monkeys"
amonkeys_qual$Song <- "505"
amonkeys_qual <- amonkeys_qual[, c(ncol(amonkeys_qual) - 1, ncol(amonkeys_qual), 1:(ncol(amonkeys_qual) - 2))]walters_qual <- get_track_audio_features("4SqWKzw0CbA05TGszDgMlc")
walters_qual$Artist <- "The Walters"
walters_qual$Song <- "I Love You So"
walters_qual <- walters_qual[, c(ncol(walters_qual) - 1, ncol(walters_qual), 1:(ncol(walters_qual) - 2))]orodrigo_qual <- get_track_audio_features("4ZtFanR9U6ndgddUvNcjcG")
orodrigo_qual$Artist <- "Olivia Rodrigo"
orodrigo_qual$Song <- "good 4 u"
orodrigo_qual <- orodrigo_qual[, c(ncol(orodrigo_qual) - 1, ncol(orodrigo_qual), 1:(ncol(orodrigo_qual) - 2))]Combining rows
API_extract <- rbind(hstyles_qual, ganimals_qual, slacy_qual, jharlow_qual, klaroi_qual)Reading Data extracted from Tunebat
tunebat_extracted<- read.csv("https://raw.githubusercontent.com/genmid13/data607/main/Tunebat%20Extraction.csv")
tunebat_extracted <- head(tunebat_extracted, n = nrow(tunebat_extracted))
tunebat_extracted## Artist Song Duration BPM
## 1 Harry Styles As It Was 2:47 174
## 2 Glass Animals Heat Waves 3:59 81
## 3 Steve Lacy Bad Habit 3:52 169
## 4 Jack Harlow First Class 2:54 107
## 5 The Kid Laroi STAY (feat. Justin Beiber) 2:22 170
## 6 Kate Bush Running Up That Hill (A Deal With God) 4:59 108
## 7 the Neighbourhood Sweater Weather 4:00 124
## 8 Steve Lacy Dark Red 2:53 172
## 9 Carolina Gaitan We Don’t Talk About Bruno 3:36 206
## 10 Lil Nas X Industry Baby (feat. Jack Harlow) 3:32 150
## 11 Joji Glimpse of Us 3:53 170
## 12 J.Cole No Role Modelz 4:53 100
## 13 Kodak Black Super Gremlin 3:21 73
## 14 Drake Knife Talk (with 21 Savage ft. Project Pat 4:03 146
## 15 Future WAIT FOR U (feat. Drake & Tems) 3:10 83
## 16 Lizzo About Damn Time 3:12 109
## 17 Imagine Dragons Enemy (with JID) 2:53 77
## 18 Arctic Monkeys 505 4:14 140
## 19 The Walters I Love You So 2:40 76
## 20 Olivia Rodrigo good 4 u 2:58 167
## Release.Date Explicit
## 1 3/31/22 No
## 2 8/7/20 No
## 3 6/29/22 Yes
## 4 5/6/22 Yes
## 5 7/9/21 Yes
## 6 9/16/85 No
## 7 4/19/13 No
## 8 2/20/17 No
## 9 11/19/21 No
## 10 7/23/21 Yes
## 11 6/10/22 No
## 12 12/9/14 Yes
## 13 10/30/21 Yes
## 14 9/3/21 Yes
## 15 4/29/22 Yes
## 16 4/14/22 Yes
## 17 10/28/21 No
## 18 4/24/07 No
## 19 11/28/14 No
## 20 5/21/21 Yes
## Album
## 1 Single
## 2 Dreamland
## 3 Single
## 4 Come Home The Kids Miss Your
## 5 Single
## 6 Hounds Of Love
## 7 I Love You.
## 8 Single
## 9 Encanto (Original Motion Picture Soundtrack)
## 10 Single
## 11 Single
## 12 2014 Forest Hills Drive
## 13 Sniper Gang Presents Syko Bob & Snapkatt: Nightmare Babies
## 14 Certified Lover Boy
## 15 I NEVER LIKED YOU
## 16 Single
## 17 Single
## 18 Favourite Worst Nightmare (Standard Version)
## 19 Single
## 20 SOUR
## Label
## 1 Columbia
## 2 Polydor Records
## 3 L-M Records/RCA Records
## 4 Genertion Now/Atlantic
## 5 Columbia
## 6 Fish People
## 7 Columbia
## 8 Three Quarter
## 9 Walt Disney Records
## 10 Columbia
## 11 88rising Music/Warner Records
## 12 Roc Nation Records LLC
## 13 Atlantic/Sniper Gang
## 14 OVO
## 15 Epic/Freebandz
## 16 Nice Life/Atlantic
## 17 KIDinaKORNER/Interscope Records
## 18 Domino/Warner Records
## 19 Warner Records
## 20 Olivia Rodrigo PSdata <- cbind(tunebat_extracted, API_extract)
data <- data[, -c(9,10)]
data <- data %>%
select(Artist, Song, duration_ms, BPM, Release.Date, Explicit, Album, Label, danceability, energy, key, loudness, mode, Song, speechiness, acousticness, instrumentalness, liveness, valence, tempo, tempo, time_signature)
data$duration_ms <- round(data$duration_ms, -3) # Round to the nearest thousand
data$duration_ms <- substr(data$duration_ms, 1, 3) # Keep only the first three numbers
data## Artist Song duration_ms BPM
## 1 Harry Styles As It Was 167 174
## 2 Glass Animals Heat Waves 239 81
## 3 Steve Lacy Bad Habit 173 169
## 4 Jack Harlow First Class 174 107
## 5 The Kid Laroi STAY (feat. Justin Beiber) 142 170
## 6 Kate Bush Running Up That Hill (A Deal With God) 167 108
## 7 the Neighbourhood Sweater Weather 239 124
## 8 Steve Lacy Dark Red 173 172
## 9 Carolina Gaitan We Don’t Talk About Bruno 174 206
## 10 Lil Nas X Industry Baby (feat. Jack Harlow) 142 150
## 11 Joji Glimpse of Us 167 170
## 12 J.Cole No Role Modelz 239 100
## 13 Kodak Black Super Gremlin 173 73
## 14 Drake Knife Talk (with 21 Savage ft. Project Pat 174 146
## 15 Future WAIT FOR U (feat. Drake & Tems) 142 83
## 16 Lizzo About Damn Time 167 109
## 17 Imagine Dragons Enemy (with JID) 239 77
## 18 Arctic Monkeys 505 173 140
## 19 The Walters I Love You So 174 76
## 20 Olivia Rodrigo good 4 u 142 167
## Release.Date Explicit
## 1 3/31/22 No
## 2 8/7/20 No
## 3 6/29/22 Yes
## 4 5/6/22 Yes
## 5 7/9/21 Yes
## 6 9/16/85 No
## 7 4/19/13 No
## 8 2/20/17 No
## 9 11/19/21 No
## 10 7/23/21 Yes
## 11 6/10/22 No
## 12 12/9/14 Yes
## 13 10/30/21 Yes
## 14 9/3/21 Yes
## 15 4/29/22 Yes
## 16 4/14/22 Yes
## 17 10/28/21 No
## 18 4/24/07 No
## 19 11/28/14 No
## 20 5/21/21 Yes
## Album
## 1 Single
## 2 Dreamland
## 3 Single
## 4 Come Home The Kids Miss Your
## 5 Single
## 6 Hounds Of Love
## 7 I Love You.
## 8 Single
## 9 Encanto (Original Motion Picture Soundtrack)
## 10 Single
## 11 Single
## 12 2014 Forest Hills Drive
## 13 Sniper Gang Presents Syko Bob & Snapkatt: Nightmare Babies
## 14 Certified Lover Boy
## 15 I NEVER LIKED YOU
## 16 Single
## 17 Single
## 18 Favourite Worst Nightmare (Standard Version)
## 19 Single
## 20 SOUR
## Label danceability energy key loudness mode
## 1 Columbia 0.520 0.731 6 -5.338 0
## 2 Polydor Records 0.761 0.525 11 -6.900 1
## 3 L-M Records/RCA Records 0.603 0.784 6 -4.023 1
## 4 Genertion Now/Atlantic 0.902 0.582 5 -5.902 0
## 5 Columbia 0.591 0.764 1 -5.484 1
## 6 Fish People 0.520 0.731 6 -5.338 0
## 7 Columbia 0.761 0.525 11 -6.900 1
## 8 Three Quarter 0.603 0.784 6 -4.023 1
## 9 Walt Disney Records 0.902 0.582 5 -5.902 0
## 10 Columbia 0.591 0.764 1 -5.484 1
## 11 88rising Music/Warner Records 0.520 0.731 6 -5.338 0
## 12 Roc Nation Records LLC 0.761 0.525 11 -6.900 1
## 13 Atlantic/Sniper Gang 0.603 0.784 6 -4.023 1
## 14 OVO 0.902 0.582 5 -5.902 0
## 15 Epic/Freebandz 0.591 0.764 1 -5.484 1
## 16 Nice Life/Atlantic 0.520 0.731 6 -5.338 0
## 17 KIDinaKORNER/Interscope Records 0.761 0.525 11 -6.900 1
## 18 Domino/Warner Records 0.603 0.784 6 -4.023 1
## 19 Warner Records 0.902 0.582 5 -5.902 0
## 20 Olivia Rodrigo PS 0.591 0.764 1 -5.484 1
## speechiness acousticness instrumentalness liveness valence tempo
## 1 0.0557 0.3420 1.01e-03 0.3110 0.662 173.930
## 2 0.0944 0.4400 6.70e-06 0.0921 0.531 80.870
## 3 0.0620 0.4460 8.32e-06 0.1190 0.769 172.041
## 4 0.1090 0.1110 3.18e-06 0.1110 0.332 107.005
## 5 0.0483 0.0383 0.00e+00 0.1030 0.478 169.928
## 6 0.0557 0.3420 1.01e-03 0.3110 0.662 173.930
## 7 0.0944 0.4400 6.70e-06 0.0921 0.531 80.870
## 8 0.0620 0.4460 8.32e-06 0.1190 0.769 172.041
## 9 0.1090 0.1110 3.18e-06 0.1110 0.332 107.005
## 10 0.0483 0.0383 0.00e+00 0.1030 0.478 169.928
## 11 0.0557 0.3420 1.01e-03 0.3110 0.662 173.930
## 12 0.0944 0.4400 6.70e-06 0.0921 0.531 80.870
## 13 0.0620 0.4460 8.32e-06 0.1190 0.769 172.041
## 14 0.1090 0.1110 3.18e-06 0.1110 0.332 107.005
## 15 0.0483 0.0383 0.00e+00 0.1030 0.478 169.928
## 16 0.0557 0.3420 1.01e-03 0.3110 0.662 173.930
## 17 0.0944 0.4400 6.70e-06 0.0921 0.531 80.870
## 18 0.0620 0.4460 8.32e-06 0.1190 0.769 172.041
## 19 0.1090 0.1110 3.18e-06 0.1110 0.332 107.005
## 20 0.0483 0.0383 0.00e+00 0.1030 0.478 169.928
## time_signature
## 1 4
## 2 4
## 3 4
## 4 4
## 5 4
## 6 4
## 7 4
## 8 4
## 9 4
## 10 4
## 11 4
## 12 4
## 13 4
## 14 4
## 15 4
## 16 4
## 17 4
## 18 4
## 19 4
## 20 4data$duration_ms <- as.numeric(data$duration_ms)
mean_duration <- mean(data$duration_ms, na.rm = TRUE)
cat("The average duration of the most streamed songs in 2022 is", mean_duration, "seconds with songs ranging", sd(data$duration_ms, na.rm = TRUE), "seconds below or above the mean")## The average duration of the most streamed songs in 2022 is 179 seconds with songs ranging 33.00399 seconds below or above the meanggplot(data, aes(x = duration_ms)) +
geom_histogram(fill = "lightblue", color = "black", bins = 4) +
geom_vline(xintercept = mean_duration, color = "red", linetype = "dashed", size = 1) ## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
)
mean_tempo <- mean(data$tempo, na.rm = TRUE)
cat("The average tempo of the most streamed songs in 2022 is", mean_tempo, "with songs ranging", sd(data$temp, na.rm = TRUE), "below or above the mean")## The average tempo of the most streamed songs in 2022 is 140.7548 with songs ranging 40.1464 below or above the meanggplot(data, aes(x = tempo)) +
geom_histogram(fill = "lightblue", color = "black", bins = 10) +
geom_vline(xintercept = mean_tempo, color = "red", linetype = "dashed", size = 1) 
mean_danceability <- mean(data$danceability, na.rm = TRUE)
cat("The average danceability of the most streamed songs in 2022 is", mean_danceability, "with songs ranging", sd(data$danceability, na.rm = TRUE), "below or above the mean")## The average danceability of the most streamed songs in 2022 is 0.6754 with songs ranging 0.1415621 below or above the meanggplot(data, aes(x = danceability)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_danceability, color = "red", linetype = "dashed", size = 1)
mean_energy <- mean(data$energy, na.rm = TRUE)
cat("The average energy of the most streamed songs in 2022 is", mean_energy, "with songs ranging", sd(data$energy, na.rm = TRUE), "below or above the mean")## The average energy of the most streamed songs in 2022 is 0.6772 with songs ranging 0.1066848 below or above the meanggplot(data, aes(x = energy)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_energy, color = "red", linetype = "dashed", size = 1)
mean_key <- mean(data$key, na.rm = TRUE)
cat("The average key of the most streamed songs in 2022 is", mean_key, "with songs ranging", sd(data$key, na.rm = TRUE), "below or above the mean")## The average key of the most streamed songs in 2022 is 5.8 with songs ranging 3.270281 below or above the meanggplot(data, aes(x = key)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_key, color = "red", linetype = "dashed", size = 1)
mean_loudness <- mean(data$loudness, na.rm = TRUE)
cat("The average loudness of the most streamed songs in 2022 is", mean_loudness, "with songs ranging", sd(data$loudness, na.rm = TRUE), "below or above the mean")## The average loudness of the most streamed songs in 2022 is -5.5294 with songs ranging 0.9542494 below or above the meanggplot(data, aes(x = loudness)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_loudness, color = "red", linetype = "dashed", size = 1)
mean_speechiness <- mean(data$speechiness, na.rm = TRUE)
cat("The average speechiness of the most streamed songs in 2022 is", mean_speechiness, "with songs ranging", sd(data$speechiness, na.rm = TRUE), "below or above the mean")## The average speechiness of the most streamed songs in 2022 is 0.07388 with songs ranging 0.02419425 below or above the meanggplot(data, aes(x = speechiness)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_speechiness, color = "red", linetype = "dashed", size = 1)
mean_acousticness <- mean(data$acousticness, na.rm = TRUE)
cat("The average acousticness of the most streamed songs in 2022 is", mean_acousticness, "with songs ranging", sd(data$acousticness, na.rm = TRUE), "below or above the mean")## The average acousticness of the most streamed songs in 2022 is 0.27546 with songs ranging 0.1740398 below or above the meanggplot(data, aes(x = acousticness)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_acousticness, color = "red", linetype = "dashed", size = 1)
mean_instrumentalness <- mean(data$instrumentalness, na.rm = TRUE)
cat("The average instrumentalness of the most streamed songs in 2022 is", mean_instrumentalness, "with songs ranging", sd(data$instrumentalness, na.rm = TRUE), "below or above the mean")## The average instrumentalness of the most streamed songs in 2022 is 0.00020564 with songs ranging 0.0004126385 below or above the meanggplot(data, aes(x = instrumentalness)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_instrumentalness, color = "red", linetype = "dashed", size = 1)
mean_liveness <- mean(data$liveness, na.rm = TRUE)
cat("The average liveness of the most streamed songs in 2022 is", mean_liveness, "with songs ranging", sd(data$liveness, na.rm = TRUE), "below or above the mean")## The average liveness of the most streamed songs in 2022 is 0.14722 with songs ranging 0.08451193 below or above the meanggplot(data, aes(x = liveness)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_liveness, color = "red", linetype = "dashed", size = 1)
mean_valence <- mean(data$valence, na.rm = TRUE)
cat("The average valence of the most streamed songs in 2022 is", mean_valence, "with songs ranging", sd(data$valence, na.rm = TRUE), "below or above the mean")## The average valence of the most streamed songs in 2022 is 0.5544 with songs ranging 0.1545637 below or above the meanggplot(data, aes(x = valence)) +
geom_histogram(fill = "lightblue", color = "black", bins = 3) +
geom_vline(xintercept = mean_valence, color = "red", linetype = "dashed", size = 1)
avg_qual <- data.frame(mean_duration, mean_tempo, mean_danceability, mean_energy, mean_key, mean_loudness, mean_speechiness, mean_acousticness, mean_instrumentalness, mean_liveness, mean_valence)
avg_qual <- pivot_longer(avg_qual, everything(), names_to = "Qualities", values_to = "Average")
avg_qual$Qualities <- c("Duration (in seconds)", "Tempo", "Danceability", "Energy", "Key", "Loudness", "Speechiness", "Acousticness", "Instrumentalness", "Liveness", "Valence")
kable(avg_qual, format = "html", caption = "Summary Table") %>%
kable_styling("striped", full_width = FALSE)| Qualities | Average |
|---|---|
| Duration (in seconds) | 179.0000000 |
| Tempo | 140.7548000 |
| Danceability | 0.6754000 |
| Energy | 0.6772000 |
| Key | 5.8000000 |
| Loudness | -5.5294000 |
| Speechiness | 0.0738800 |
| Acousticness | 0.2754600 |
| Instrumentalness | 0.0002056 |
| Liveness | 0.1472200 |
| Valence | 0.5544000 |
Conclusion
Intro (Theme: Summer Experience) (optional) If used, the intro should be the shortest section while using the most non-filler language. Most of the language should only be repeated once here.
Verse 1 (Theme: Romance//Love) The first verse the second longest section where only 1/3 should use non-filler language. However, most of that language should be distinct.
Chorus (theme: Prioritizing Relationships) The chorus should be longer than the intro/outro but less than the verses. The chorus has most filler language and the least distinctiveness in its non-filler language
Verse 2 (Theme: Temporal Longing) The second verse has the most words used. Almost all of it non-filler language (that only take 1/3 of the section) is distinct.
Outro (Theme: Summer Vacation) (optional) The outro uses more words than the intro (and less than the others) but is less complex in its non-filler language.