UNIVERSITY PARK, Pa. — A team of Penn State faculty, librarians, and undergraduate students was recently selected for Phase One of the $100,000 Nittany AI Challenge for their prototype that uses artificial intelligence to automate parts of the University Libraries’ digital badge program. The group will demonstrate their prototype during Penn State Startup Week on March 27 in hopes of receiving an additional $5,000 to create a minimum viable product.
The challenge, a second-year initiative from the Penn State EdTech Network, asks teams to develop artificial intelligence-based solutions that improve the student experience at Penn State, solve real problems that the University is facing, and generate innovative startup ideas.
The team’s submission, “From Micro to Macro: Applying Machine Learning to Scale Up Competency Based Learning at PSU,” proposes using AI to evaluate information literacy skills that students work to improve by completing digital badges. Information literacy is the set of integrated abilities encompassing the reflective discovery of information, the understanding of how information is produced and valued, and the use of information in creating new knowledge and participating ethically in communities of learning.
Specifically, the group proposes that the use of automated essay scoring (AES) can lessen the time needed to evaluate students’ responses to digital badge submissions. AES is an existing approach used to assess the quality, accuracy and relevancy of natural language writing. Badges provide an opportunity to evolve the AES field to both improve student experience by providing instantaneous feedback, while also reducing the load on human graders by providing them with automated analysis of the badge steps that they are grading.
Digital badges, also known as micro-credentials, serve as visual representations of different skills learned, allowing users to have a record of the various competencies they have acquired. Digital badges have been used in gaming and online scenarios to motivate behavior and establish credibility, but now organizations are using them in learning environments to represent learned skills and knowledge. Badges allow users to guide their own learning and employers to recognize education that happens outside the classroom.
“The University Libraries created digital badges in 2013 because we are really invested in information literacy skills for students, but we had an unbalanced way to reach them,” said Emily Rimland, information literacy librarian and learning technologies coordinator with the University Libraries and one of the team leads. “Badges allow us to reach different students in unique ways and provide them with a linear pathway to learn new skills.”
In addition to Emily Rimland, the team includes Victoria Raish, online learning librarian with the University Libraries; Jeff Rimland, assistant teaching professor of information sciences and technology; and IST undergraduate students Angela Demarco and Arianna Scheidell. The group’s proposal was one of ten selected from 71 submissions for further participation in the challenge.
In the Libraries’ existing program, librarians create badges using the University’s digital badge platform that feature a series of steps for the learner to complete. Over the past 18 months, the badges have gained immense popularity across the University’s general education courses and have dramatically increased the time required to provide meaningful and personalized feedback to the learner. In the fall 2017 semester alone, Penn State librarians evaluated over 10,000 badge steps.
The group thinks that artificial intelligence could be the solution.
“Due to a recent surge in adoption of digital badges at Penn State, the University Libraries was looking to AI to help with aspects of evaluating information literacy badges so that they could be offered at quality, speed, and scale for additional courses,” the group wrote in their proposal.
By using AI to help analyze student evidence submitted to earn a badge, the group believes that they can increase information literacy instruction and engage more students with this crucial skill.
“Our digital badge program has flourished, and we would like to invite more courses to use it. However, we need a more sustainable solution with regards to evaluation of student responses,” the group’s proposal states. “AI could be used to triage students’ work in a variety of ways such as giving an initial quality score, highlighting strengths and weaknesses, and flagging plagiarized responses in order to streamline the evaluation process.” This approach could also be more widely applied in other learning environments such as discussion boards.
After the Libraries piloted a badge program in 2016 for students for certain classes offered through Penn State World Campus, faculty noted the students’ improved quality of sources and more accurate citations, leading to a better academic output. And though the process ideally requires human intervention to build relationships and personal connections to Penn State, artificial intelligence could lighten the load for evaluators and allow more courses to participate.
“The Nittany AI Challenge is an opportunity for a cross-disciplinary partnership between the University Libraries and the College of Information Sciences and Technology, which fits nicely within the University's Strategic Plan themes of transforming education and driving digital innovation,” said Jeff Rimland. “This project has the potential to springboard a solution to a problem, one we had identified as having broad implications at Penn State and beyond. We hope to harness the power of AI to be able to offer badges on a scale befitting their potential impact.”
Student participation in this challenge is also a benefit for both the teams and the students themselves. “I love the opportunity to work on these more challenging problems with faculty, and I feel that I’m learning a lot more from this hands-on activity than I ever would in the classroom,” said Angie Demarco.
“I believe that AI will fundamentally transform how the world works,” added Arianna Scheidell. “Right now, we only see a glimpse of what it can do, but that's why working at the cutting edge is so exciting.”
The remaining teams in the Nittany AI Challenge will demonstrate their prototypes from noon to 5 p.m. on Tuesday, March 27, in 134 HUB-Robeson Center. This event is free and open to the public.