[Air-L] ACM RecSys Challenge Kick-off

[Starke, Alain]

Dear all,

I am pleased to announce the kick-off of the 'RecSys Challenge', which is part of the 19th ACM Conference on Recommender Systems, Prague, September 22-26.
The RecSys (Recommender Systems) Challenge is the official challenge of the ACM RecSys conference series, in which participants need to come up with the best approach for a given problem with a dataset. The challenge has been organized since 2012 and each a different company makes a dataset available: https://www.recsyschallenge.com/2025/.

This year's challenge features the concept of Universal Behavioral Profiles-user representations that encode essential aspects of each individual's past interactions. These profiles are designed to be universally applicable across multiple predictive tasks, such as churn prediction and product recommendations. By developing representations that capture fundamental patterns in user behavior, we enable models to generalize effectively across different applications.

The challenge involves to predict a data model on several tasks, including Churn Prediction, Product Propensity and Category Propensity. There are also several hidden tasks. The data will be evaluated with the AUROC metric. All information is available on the challenge website: https://www.recsyschallenge.com/2025/.

Prizes
The first three teams that come up with the best model can win prizes of $3,500, $2,500 and $1,500. The best academic teams can win prizes of $2,500, $1,500, and $500.
The dataset has been released on March 10, the first leaderboard will be published on April 10.

Check out all details on the website: https://www.recsyschallenge.com/2025/.

Happy participation!

Kind regards,
Alain Starke
Publicity Co-Chair ACM RecSys