by Claudia Shao, Varshitha Chennamsetti, Winnie Zheng
With over 430 million active listeners, how can Spotify personalize recommendations that cater to users’ varying music tastes? To help recommend the right songs and keep users streaming for longer hours on the app, we built a user behavior prediction model using methods like Decision Tree, SVM, and RNN to predict whether a song will be skipped.