Announcing the Nittany AI Challenge Finalists

SHARE THIS POST

Artificial intelligence (AI) is everywhere. It creates carpooling routes, filters spam from your email and can even optimize your thermostat use. The Nittany AI Challenge, hosted by Penn State University and co-sponsored by Blackboard, engages students to develop innovative uses of AI to tackle real problems on campus and in the world.

Recently, I had the privilege of representing Blackboard on a judging panel as part of the challenge. The purpose of the panel was to judge 11 groups based on their proofs of concept, demonstrations and roadmaps. We selected the best five to continue to the next stage of the challenge.

Now, the selected teams will have several months to turn their proofs of concept into a minimal viable product and compete for additional seed funding for their product. Many of the teams submitted strong ideas that have a real chance at impacting students and faculty at Penn State or even well beyond college campuses or academia. Here’s a quick rundown on the five finalists:

  • Pathfinder: Pathfinder uses machine learning to identify successful course sign-up patterns. With it, students can select a course and see course “paths”. These paths fulfill prerequisites for the course and aim to improve students’ readiness for their chosen courses. Results will let users know that, for instance, students like them who take BIO 401 average a B-, but those who first take BIO 342 average an A- in the same course. This way, the app can guide students through a maze of courses.
  • LionPlanner: LionPlanner also focuses on curriculum planning. It uses student interests and feedback to suggest not only courses, but also clubs and activities. It produces valid registrations and provides warnings and feedback about prerequisites still needed.
  • Aspire: Aspire aims to take a big picture look at not only course selection, but career planning. The basis of its suggestions is data scraped from the internet and submitted by users. Data includes resumes and LinkedIn profiles. The idea is to help users see what courses and career moves will advance them from being students to having their dream jobs.
  • ProFound: ProFound is a slick app that also addresses a real need at many institutions. At many institutions, each department maintains its own faculty directories and bios. As a result, the information they contain is rarely coordinated and often has significant gaps. ProFound combs the web to find information about professors and uses AI to identify which bits of information mean what. The app then builds a bio page and a preview for each identified professor. With it, students can search for professors by specialty, course offerings, academic research, and more.
  • Micro to Macro (FMTM): FMTM uses machine learning to check student text responses. The library system at PSU has built a successful micro-credentialing program to teach research skills. This has created a problem: the responses can take hours to wade through. The idea is to use the app to pre-screen responses that are likely to be irrelevant, spam, or inadequate.

Each of the winning teams was impressive. Their understanding of AI techniques was intelligent and thorough. Their applications were serious and useful. We chose them for their good ideas, exhibited ability to execute and clear paths forward for both features and user experience. It will be exciting to see what they cook up in the next few months-and we’ll keep you posted on their progress.