MIND News Recommendation Competition Forum

Go back to competition Back to thread list Post in this thread

> On what dataset, should we compute our predictions?

Hello dear organizers,

I am afraid, I missed the dataset on which one should compute the predictions? On submission tab, it is mentioned: "Participants need to submit the ranking results of news in each impression generated by a recommender system." But, where are these impressions? In the validation set? Or in the training set?

Posted by: learner @ Aug. 4, 2020, 5:37 a.m.
Post in this thread