Development of an AI model for possible analysis of students’ concentration and understanding levels in online lectures for the students of faculty and graduate school of Department of Education

Proposal of AI utilization method was praised in descriptive evaluation.

Received gold award and excellence award in “AI leaning data hackathon”

[February 15, 2022]

<From the left, JOO Jeong-hun (Department of Education) and PARK Kang-yoon and KIM Kil-jae (master course of graduate school)>

“What role will AI be able to play in smart education environment?”

 “AI utilization method” in the area of teaching/learning method proposed by YU students (President CHOI Oe-chool) attracts attention.

  PARK Kang-yoon (3rd grade of master course, Department of Education, graduate school), JOO Jeong-hun (3rd grade of Department of Education), and KIM Kyu-ri (3rd grade of Department of Visual Design) developed an AI model to identify understanding and concentration levels of students in online lectures. They developed an AI eye-tracing model based on eyeball data and developed indexes of fixation point data collecting, understanding, and concentrating levels. JOO Jeong-hun said, “The developed model may help teachers to identify concentration and understanding levels of students in the lessons using online lectures. It may be used as a solution for learning deficiency issues that occur in online education.”

  Recently, online education has become a topic due to prolonged expansion of COVID-19. With rapidly increased demand for intact online education, various types of learning deficiency occurs and it attracts attention as a supplementary measure for students who are demanders of education. The idea received gold award in “2021 Eyeball movement video AI learning data hackathon” held by Ministry of Information & Communication and National Information Society Agency.

  Also, PARK Kang-yoon and KIM Kil-jae (1st grade of master curse of graduate school) accessed to AI utilization method in the viewpoint of education evaluation. They took attention in recent situation that expansion of descriptive and essay-type evaluation systems have become an issue, developed AI similar with the level of human graders, and proposed an AI model to reduce burden of work of teachers in case of descriptive evaluation. KIM Kil-jae said, “It is inefficient to evaluate the essays of all students through human due to consumption of huge costs and time. The constructed AI evaluation model will help teachers to enhance efficiency of education.” The AI model constructed by them received excellence award in “essay data AI lessening data hackathon” held last year.

  PARK Kang-yoon who was included in the winner lists of two contests due to creative ideas commented, “I think that diverse challenges to use AI will make a foundation for construction of smart education environment and virtuous cycle of data in future. I am conducting in-depth researches on the method to enhance quality of education using diverse deep learning and machine running algorithms.”