Cookies help us deliver our services. By using our services, you agree to our use of cookies. Learn more

crowdAI is shutting down - please read our blog post for more information

Spotify Sequential Skip Prediction Challenge

Predict if users will skip or listen to the music they're streamed


Ground truth

Posted by zb over 1 year ago

in the overview section, it’s stated that “For this challenge we use the skip_2 field of the session logs as our ground truth.”
Doing some EDA, there are plenty of rows whereby skip_2 is true, but eventually the not_skipped column is False, which is not a representation of whether or not a track was skipped. Shouldnt the not_skipped column be our target column? thanks!

Posted by brianbrost  over 1 year ago |  Quote

Hi @zb, I’m not sure exactly what you mean, but the fact that skip_2 is true should generally imply that not_skipped is false. Regarding the target column, we could have used any of the skip columns, but the skip_2 column is interesting itself, and is also quite balanced between true and false.