The Inaugural 2020 RMP Data-Driven Research Challenge
The INFORMS/Revenue Management & Pricing (RMP) Section and NetEase Cloud Music are partnering to offer RMP members access to NetEase Cloud Music App's impression-level data to encourage data-driven research on innovative marketplaces.
In this competition, researchers will compete by building data-driven models using real data to address either some of the suggested questions below or questions of their own interest.
Eligibility Criteria: Who can enter?
(1) RMP members are encouraged to participate.
(2) At least one author of the submitted paper must be an existing RMP member in 2020.
(3) Each RMP member may enter the competition by submitting at most one paper.
Judging Criteria: What the judges will be looking for?
All entries will be judged according to the following criteria
Criteria 1: Data-Driven.
Criteria 2: Potential Impact on Practice.
Criteria 3: Contribution to the Research Literature.
Criteria 4: Generalizability / Scalability.
April 1st, 2021: Competition submissions deadline.
May 2021: Judges to review submissions and select finalists.
June 2021 during the RMP Section 2021 Conference (exact date -- TBA): Finalist presentations, judges select winners, and winner(s) announcement.
* Finalists can opt to have their reviews by judges forwarded to the Revenue Management and Market Analytics department at Management Science if they choose to make a submission to that department.
* Finalist will be invited for a Fast Track submission to the Revenue Management and Marketplace Design department at Naval Research Logistics. Fast Track means that the paper will go through 1 round of review before making a final decision.
Data Acquisition and Submission Guidelines:
(1) You need to be an RMP member in 2020. See how to join.
(2) You need to use your INFORMS Member ID and Password to access the data.
(3) You can access the data here.
(4) Before the deadline, you can submit the paper to RMPDataCompetition2020@gmail.com with the subject "2020 Data-Driven Research Challenge".
* If you have any further questions, please email one of the co-chairs.
NetEase Cloud Music would like to encourage researchers to explore the provided data and develop innovative solutions to address the following problems (or other research problems of their own choosing):
- The company defines a user to be inactive if she has a zero or very low average click probability on recommended cards. How could the company differentiate active users from inactive users? How could the company design the recommender system to minimize the number of inactive users?
- What are the characteristics and preferences of active users on the platform? How could the platform predict whether a user will be active or inactive from her early-on actions, such as clicks, likes, and shares? How could the platform design the recommender system to maximize the number of active users?
- How do different types of feedback information, such as the number of likes and follows, change a creator's motivation to publish new content? How could the platform design the recommender system to encourage creators to create more content?
- The company's long-term goal on the tab is to maximize the daily number of clicks/plays and the daily number of content created. How could the company create a recommender system that trades off short-term goals, such as the number of clicks/plays in one day, with this long-term goal?