Data Science Competition

Days Held

Phase 1: Saturday–Sunday, October 30–31, 2021
Phase 2: Saturday–Sunday, November 6–7, 2021

The Data Science Competition is being added to SC this year to provide students with an opportunity to showcase their data skills and the problems they are passionate about solving while integrating additional educational paths to HPC. Details of the competition are still being planned out, so this description may change as these plans evolve, but should be finalized soon. The teams will compete in two phases prior to the actual conference and present their projects to be judged at the end. Each phase will happen over a weekend. The teams will present their approaches and projects to the judges at the end of the competition. After judging, winning teams will be announced at the Students@SC Awards ceremony. Teams will also be asked to share their work publicly.

Preparing for the Competition

Students will be expected to have some relevant experience or background, but are not required to be experts. Once accepted, teams can attend various workshops and webinars to help them prepare prior to the competition. These will cover the skills and tools they’ll need to be successful, including an introduction to the cloud platform hosting the competition.

Phase 1

The first phase of the competition will be to approach the same mystery data set and general problem scope provided to the teams by the committee. This will be revealed at the start of the competition. Each team will be expected to derive some insights and present them to the judges. This phase will also allow teams to work through any hiccups in working on the cloud platform, teamwork, scale, time management, etc.

Phase 2

During the week between phases students will come up with a short proposal for an independent project, including the proposed data source(s), which will be approved by the committee before phase two. The second phase will have teams approach this independent project. Teams are encouraged to focus on problems relevant to St. Louis this year, but the categories they can explore are broad, ranging from areas such as education, climate or healthcare problems. This phase will be more intensive than Phase 1 and students will prepare a presentation that emcompasses everything from scoping the problem and relevant background, to their approach, and their final insights.

Teams & Process

Teams are composed of 4-5 students and a mentor. Students are encouraged to form interdisciplinary teams that bridge the skills required of successful many data science projects. This can include many disciplines that teach and utilize data, analytics, and research skills. This includes students outside of traditional routes to data science like computer science or machine learning. Research using data skills span many areas from biology to linguistics to physics and more. A team mentor is optional, but encouraged to apply with the team and is responsible for providing basic guidance and helping troubleshoot issues during the competition. Students will work together to define their datasets, problems, and approaches.

Applications are open to undergraduate students and first year graduate students that are 18+ at the time of the competition start.

Support Provided

Selected teams will receive conference registration for the Technical Program and Workshops for each student member. Travel and accommodations are not provided as the competition will be completely virtual. However, students are encouraged to apply to become a student volunteer, as participation dates do not overlap and partial travel support to augment in-person attendance to SC is available to student volunteers.

As the competition is part of the Students@SC program, students can also participate in Mentor–Protégé Matching and the Job Fair.

Thank You Data Science Competition Supporter