Capstone_Markdown
Britney Llewellyn
2024-04-15
library(readr)
clean_taylor_swift_spotify_data <- read_csv("~/Working with data frames/clean_taylor_swift_spotify_data.csv")## Rows: 530 Columns: 15
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (4): album, release_date, id, track_name
## dbl (11): acousticness, danceability, energy, instrumentalness, liveness, lo...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.colnames(clean_taylor_swift_spotify_data)## [1] "album" "release_date" "id" "track_name"
## [5] "acousticness" "danceability" "energy" "instrumentalness"
## [9] "liveness" "loudness" "speechiness" "tempo"
## [13] "valence" "popularity" "duration_ms"library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(knitr)
top_3_popularity <- select(clean_taylor_swift_spotify_data, track_name,popularity)%>%
arrange(desc(popularity))%>%
head(n=3)
kable(head(top_3_popularity, 3))| track_name | popularity |
|---|---|
| Cruel Summer | 96 |
| Lover | 90 |
| Anti-Hero | 89 |
spotify screenshot
clean_distinct_spotify_data <- (clean_taylor_swift_spotify_data)%>%
arrange(desc(popularity))
clean_distinct_spotify_data <- distinct(clean_distinct_spotify_data, track_name, .keep_all = TRUE)popularity <- clean_distinct_spotify_data$popularity
correlation_with_popularity <- cor(clean_distinct_spotify_data[, sapply(clean_distinct_spotify_data, is.numeric)], popularity)
print(correlation_with_popularity)## [,1]
## acousticness -0.056909062
## danceability 0.140757318
## energy 0.001182703
## instrumentalness 0.041195021
## liveness -0.293730395
## loudness -0.079461432
## speechiness -0.037737809
## tempo -0.046831939
## valence -0.015328652
## popularity 1.000000000
## duration_ms -0.102930269popularity_vs_liveness <- clean_distinct_spotify_data %>%
mutate(popularity_group = cut(popularity, breaks = 10))
library(ggplot2)
ggplot(data = popularity_vs_liveness)+
geom_point(mapping=aes(x=popularity_group, y=liveness))+
labs(x= "Grouped Popularity", y= "Liveness")+
ggtitle("Scatter Plot of Grouped Popularity vs. Liveness")
clean_distinct_spotify_data %>%
group_by (album) %>%
summarise(mean_popularity = mean(popularity)) %>%
arrange(desc(mean_popularity))## # A tibble: 27 × 2
## album mean_popularity
## <chr> <dbl>
## 1 reputation 79.3
## 2 Lover 79.3
## 3 Midnights 78.1
## 4 1989 (Taylor's Version) 77.4
## 5 folklore 76.3
## 6 Midnights (The Til Dawn Edition) 74.7
## 7 1989 (Taylor's Version) [Deluxe] 74
## 8 evermore 73.7
## 9 Speak Now (Taylor's Version) 73.3
## 10 Red (Taylor's Version) 70.3
## # ℹ 17 more rows# Create a new dataset for continued analysis
top_5_albums <- clean_distinct_spotify_data %>%
filter(album %in% c("reputation","Lover","Midnights","1989 (Taylor's Version)", "folklore"))library(dplyr)
# Define a function to calculate correlation for each group
calculate_correlation <- function(df) {
correlation_with_popularity <- cor(df$popularity, df[, sapply(df, is.numeric)])
return(data.frame(correlation_with_popularity))
}
# Group the data by album and calculate correlation for each group
correlation_by_album <- top_5_albums %>%
group_by(album) %>%
group_modify(~ calculate_correlation(.))
# Print the results and show all values
options(tibble.width = Inf)
print(correlation_by_album)## # A tibble: 5 × 12
## # Groups: album [5]
## album acousticness danceability energy instrumentalness
## <chr> <dbl> <dbl> <dbl> <dbl>
## 1 1989 (Taylor's Version) -0.0945 0.195 -0.0791 0.288
## 2 Lover -0.427 -0.272 0.570 0.0714
## 3 Midnights -0.498 0.253 0.690 -0.282
## 4 folklore -0.370 -0.0377 0.343 -0.0394
## 5 reputation -0.216 0.239 0.0891 0.233
## liveness loudness speechiness tempo valence popularity duration_ms
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 -0.0454 -0.225 -0.125 -0.485 -0.231 1 -0.217
## 2 -0.283 0.563 -0.0740 0.0386 0.185 1 0.160
## 3 0.237 0.676 -0.193 -0.207 0.490 1 -0.345
## 4 0.379 0.626 -0.232 -0.144 0.0762 1 -0.0350
## 5 -0.311 0.210 0.0185 0.190 -0.0837 1 0.151#Function for first graph
top_5_albums %>%
filter(album== "Lover") %>%
ggplot(aes(x=popularity, y=energy))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "Lover Tracks Energy and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#Function for second graph
top_5_albums %>%
filter(album== "Lover") %>%
ggplot(aes(x=popularity, y=loudness))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "Lover Tracks Loudness and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#Code for first Midnights graph
top_5_albums %>%
filter(album== "Midnights") %>%
ggplot(aes(x=popularity, y=energy))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "Midnights Tracks Energy and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#Code for second Midnights graph
top_5_albums %>%
filter(album== "Midnights") %>%
ggplot(aes(x=popularity, y=loudness))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "Midnights Tracks Loudness and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#Code for third Midnights graph
top_5_albums %>%
filter(album== "Midnights") %>%
ggplot(aes(x=popularity, y=valence))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "Midnights Tracks Valence and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#Code for first folklore track graph
top_5_albums %>%
filter(album== "folklore") %>%
ggplot(aes(x=popularity, y=loudness))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "folklore Tracks Loudness and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#code for second folklore graph
top_5_albums %>%
filter(album== "folklore") %>%
ggplot(aes(x=popularity, y=acousticness))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "folklore Tracks Acousticness and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#code for third folklore graph
top_5_albums %>%
filter(album== "folklore") %>%
ggplot(aes(x=popularity, y=energy))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "folklore Tracks Energy and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'
#code for fourth folklore graph
top_5_albums %>%
filter(album== "folklore") %>%
ggplot(aes(x=popularity, y=liveness))+
geom_point()+
stat_smooth(method="lm")+
labs(title = "folklore Tracks Liveness and Popularity Correlation")## `geom_smooth()` using formula = 'y ~ x'