Pixel International Conferences

Formal education systems worldwide were faced with the challenge of shifting from classroom to online e-learning platforms due to the COVID19 lockdown. The shift forced teachers to spend additional time on digital content management, as well as on individual communication to gain insight into student’s cognition and provide feedback. In programming classes, especially at the university level, communication is asynchronous more often than not [1]. Since university students learn at their own pace, they run into issues that a teacher might have anticipated, but at different times and at different rates. Because communication is asynchronous, the need for teacher’s feedback disrupts a learning pace. In this paper, the application of a knowledge-driven feedback-providing automatic-programming Artificially Intelligent Teaching Assistant (AITA) is proposed as a solution to that problem. By using cognitive maps, a teacher could manage knowledge, more precisely, the concepts, the competencies, and tasks that are a part of the programming curricula [2]. Similarly, AITA could manage metaknowledge about student’s knowledge of concepts and competencies [3] and would provide specific, direct feedback to students as the result in form of automatically generated programming code. Feedback would range from styling suggestions to fixes. This way, teachers would benefit from time savings, while students would benefit from adaptive learning mechanisms provided by AITA. This paper describes AITA’s system architecture and presents the results of pilot testing.

Keywords: programming, Covid-19, STEM education, knowledge management, adaptive learning, automatic programming.

References

1. Adnan, Muhammad, and Kainat Anwar. "Online Learning amid the COVID-19 Pandemic: Students' Perspectives." Online Submission 2.1 (2020): 45-51.
2. Mead, Jerry, et al. "A cognitive approach to identifying measurable milestones for programming skill acquisition." ACM SIGCSE Bulletin 38.4 (2006): 182-194.
3. Peña, Alejandro, Humberto Sossa, and Agustin Gutiérrez. "Cognitive maps: an overview and their application for student modeling." Computación y Sistemas 10.3 (2007): 230-250.