16.1 Data Set!

data(spotify, package = "bayesrules")

We are going to use a subset Spotify data set from #TidyTuesday:

• 44 artists -> 350 songs

spotify <- spotify |> 
  select(artist, title, popularity) |> 
  mutate(artist = fct_reorder(artist, popularity, .fun = 'mean'))

table(spotify$artist) |> hist(xlab = "Songs/artist", col = 4)

We are going to illustrate the 3 approaches seen in chapter 15:

• Complete pooling
  
• No pooling
• Partial pooling

Btw my top 3 most listening song/group from last week are:

• Dina Summer - Passion (cover)
• The Do - Anita No
• Purrrple cat - stream