illdopa - some of the attributes such as loudness are objective and need no human input, while others, such as danceability require some human training. For attributes like danceability, we will manually label a pool of songs with danceability scores that we will then use to train our machine models. --Paul
↧