Virtual On line
June 22, 2020
June 22, 2020
June 26, 2021
Diversity and NSF Grantees Poster Session
In student education, learning styles can vary wildly from one student to the next. While students should receive support tailored to their specific learning style, this type of personalized support can often not be realized due to resource constraints. This paper presents an implementation of personalized learning support utilizing a random forest machine learning model built on top of an existing narrative game environment. The existing game, Gridlock, is a domain-specific narrative game that implements metacognitive strategies to assist students in learning sequential logic design, a core topic in Computer Engineering and Computer Science. The metacognitive strategies featured in the game are Roadmap, What I Know-What I Want to Know-What I Need to Solve (KWS), and Think-Aloud-Share-Solve (TA2S). Roadmap provides students with an idea of what they have learned and what they still need to learn. KWS prompts students to remember what they already know, what they want to know, and what they are trying to solve to keep them focused and on task. TA2S encourages students to think aloud by communicating with fellow classmates to share their solutions and collaborate to solve problems. On top of existing learning strategies within the game, a random forest machine learning model is used to classify students into various categories based on their learning path. To train this model, a large dataset was generated based on previously gathered information from tests of the game as well as in-classroom observations of students playing through the game. The model was verified through multiple runs with students of varying levels of subject knowledge. As they play through the game, students are classified based on their perceived knowledge of the subject matter presented to them. From this classification, students can be provided individualized assistance in the form of tutorials, hints, prompts, or even videos of experts solving similar problems. These tailored prompts provide students with immediate feedback in their areas of difficulty, maintaining the momentum of the learning process and improving student comprehension.
Tang, Y., & Hare, R. (2020, June), A Random Forest Model for Personalized Learning in a Narrative Game Paper presented at 2020 ASEE Virtual Annual Conference Content Access, Virtual On line . 10.18260/1-2--34041
ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2020 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015