Last updated: 2024-05-22

Checks: 7 0

Knit directory: stats352-spotify/

This reproducible R Markdown analysis was created with workflowr (version 1.7.1). The Checks tab describes the reproducibility checks that were applied when the results were created. The Past versions tab lists the development history.

R Markdown file: up-to-date

Great! Since the R Markdown file has been committed to the Git repository, you know the exact version of the code that produced these results.

Environment: empty

Great job! The global environment was empty. Objects defined in the global environment can affect the analysis in your R Markdown file in unknown ways. For reproduciblity it’s best to always run the code in an empty environment.

Seed: set.seed(20240522)

The command set.seed(20240522) was run prior to running the code in the R Markdown file. Setting a seed ensures that any results that rely on randomness, e.g. subsampling or permutations, are reproducible.

Session information: recorded

Great job! Recording the operating system, R version, and package versions is critical for reproducibility.

Cache: none

Nice! There were no cached chunks for this analysis, so you can be confident that you successfully produced the results during this run.

File paths: relative

Great job! Using relative paths to the files within your workflowr project makes it easier to run your code on other machines.

Repository version: 50e1f72

Great! You are using Git for version control. Tracking code development and connecting the code version to the results is critical for reproducibility.

The results in this page were generated with repository version 50e1f72. See the Past versions tab to see a history of the changes made to the R Markdown and HTML files.

Note that you need to be careful to ensure that all relevant files for the analysis have been committed to Git prior to generating the results (you can use wflow_publish or wflow_git_commit). workflowr only checks the R Markdown file, but you know if there are other scripts or data files that it depends on. Below is the status of the Git repository when the results were generated: 


working directory clean

Note that any generated files, e.g. HTML, png, CSS, etc., are not included in this status report because it is ok for generated content to have uncommitted changes.

These are the previous versions of the repository in which changes were made to the R Markdown (analysis/spotify.Rmd) and HTML (docs/spotify.html) files. If you’ve configured a remote Git repository (see ?wflow_git_remote), click on the hyperlinks in the table below to view the files as they were in that past version.

File Version Author Date Message
Rmd 50e1f72 John Blischak 2024-05-22 wflow_publish("analysis/spotify.Rmd", delete_cache = TRUE)
html 35c8864 John Blischak 2024-05-22 Build site.
Rmd e22daa6 John Blischak 2024-05-22 Add spotify analysis

This analysis attempts to classify songs into their correct musical genre using audio features. It is inspired by the original analysis by Kaylin Pavlik (@kaylinquest) in her 2019 blog post Understanding + classifying genres using Spotify audio features.

knitr::opts_chunk$set(autodep = TRUE)
spotify <- read.csv("data/spotify.csv", stringsAsFactors = FALSE)
dim(spotify)
[1] 32833    15
head(spotify)
  genre danceability energy key loudness mode speechiness acousticness
1   pop        0.748  0.916   6   -2.634    1      0.0583       0.1020
2   pop        0.726  0.815  11   -4.969    1      0.0373       0.0724
3   pop        0.675  0.931   1   -3.432    0      0.0742       0.0794
4   pop        0.718  0.930   7   -3.778    1      0.1020       0.0287
5   pop        0.650  0.833   1   -4.672    1      0.0359       0.0803
6   pop        0.675  0.919   8   -5.385    1      0.1270       0.0799
  instrumentalness liveness valence   tempo duration_ms           artist
1         0.00e+00   0.0653   0.518 122.036      194754       Ed Sheeran
2         4.21e-03   0.3570   0.693  99.972      162600         Maroon 5
3         2.33e-05   0.1100   0.613 124.008      176616     Zara Larsson
4         9.43e-06   0.2040   0.277 121.956      169093 The Chainsmokers
5         0.00e+00   0.0833   0.725 123.976      189052    Lewis Capaldi
6         0.00e+00   0.1430   0.585 124.982      163049       Ed Sheeran
                               song
1 I Don't Care (with Justin Bieber)
2                          Memories
3                      All the Time
4                     Call You Mine
5                 Someone You Loved
6   Beautiful People (feat. Khalid)
table(spotify[, 1])

  edm latin   pop   r&b   rap  rock 
 6043  5155  5507  5431  5746  4951 
spotify <- spotify[, 1:13]

Split the data into training and testing sets. The training set should have 3/4 of the samples.

numTrainingSamples <- nrow(spotify) * 3/4
trainingSet <- sample(seq_len(nrow(spotify)), size = numTrainingSamples)
spotifyTraining <- spotify[trainingSet, ]
spotifyTesting <- spotify[-trainingSet, ]

Build classification model with decision tree from the rpart package.

library("rpart")
model <- rpart(genre ~ ., data = spotifyTraining)
plot(model, margin = 0.05)
text(model)

Version Author Date
35c8864 John Blischak 2024-05-22

Calculate prediction accuracy of the model on the training and testing sets.

predictTraining <- predict(model, type = "class")
(accuracyTraining <- mean(spotifyTraining[, 1] == predictTraining))
[1] 0.38763
predictTesting <- predict(model, newdata = spotifyTesting[, -1], type = "class")
(accuracyTesting <- mean(spotifyTesting[, 1] == predictTesting))
[1] 0.3821416

Evaluate prediction performance using a confusion matrix.

table(predicted = predictTesting, observed = spotifyTesting[, 1])
         observed
predicted  edm latin  pop  r&b  rap rock
    edm    624   164  226   72   58   79
    latin  162   494  404  261  155  133
    pop      0     0    0    0    0    0
    r&b     38    84   74  242   57   87
    rap    337   470  316  540 1008  177
    rock   360   105  377  233  103  769

How does the model compare to random guessing?

predictRandom <- sample(unique(spotifyTesting[, 1]),
                        size = nrow(spotifyTesting),
                        replace = TRUE,
                        prob = table(spotifyTesting[, 1]))
(accuracyRandom <- mean(spotifyTesting[, 1] == predictRandom))
[1] 0.1525155

sessionInfo()
R version 4.3.3 (2024-02-29 ucrt)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 11 x64 (build 22631)

Matrix products: default


locale:
[1] LC_COLLATE=English_United States.utf8 
[2] LC_CTYPE=English_United States.utf8   
[3] LC_MONETARY=English_United States.utf8
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.utf8    

time zone: America/New_York
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] rpart_4.1.23    workflowr_1.7.1

loaded via a namespace (and not attached):
 [1] jsonlite_1.8.8    highr_0.10        compiler_4.3.3    promises_1.3.0   
 [5] Rcpp_1.0.12       stringr_1.5.1     git2r_0.33.0      callr_3.7.6      
 [9] later_1.3.2       jquerylib_0.1.4   yaml_2.3.8        fastmap_1.1.1    
[13] R6_2.5.1          knitr_1.46        tibble_3.2.1      rprojroot_2.0.4  
[17] bslib_0.7.0       pillar_1.9.0      rlang_1.1.3       utf8_1.2.4       
[21] cachem_1.0.8      stringi_1.8.4     httpuv_1.6.15     xfun_0.43        
[25] getPass_0.2-4     fs_1.6.4          sass_0.4.9        cli_3.6.2        
[29] magrittr_2.0.3    ps_1.7.6          digest_0.6.35     processx_3.8.4   
[33] rstudioapi_0.16.0 lifecycle_1.0.4   vctrs_0.6.5       evaluate_0.23    
[37] glue_1.7.0        whisker_0.4.1     codetools_0.2-20  fansi_1.0.6      
[41] rmarkdown_2.26    httr_1.4.7        tools_4.3.3       pkgconfig_2.0.3  
[45] htmltools_0.5.8.1