DIGITAL LIBRARY
INTELLIGENT TOOL FOR DEVELOPING STUDENT’S SOCIAL SKILLS
1 Moscow City University (RUSSIAN FEDERATION)
2 Higher School of Economics (RUSSIAN FEDERATION)
3 ORT de Ginzburg school (RUSSIAN FEDERATION)
4 St. Petersburg branch of World ORT (RUSSIAN FEDERATION)
5 Tel Aviv University (ISRAEL)
About this paper:
Appears in: EDULEARN20 Proceedings
Publication year: 2020
Pages: 2408-2413
ISBN: 978-84-09-17979-4
ISSN: 2340-1117
doi: 10.21125/edulearn.2020.0744
Conference name: 12th International Conference on Education and New Learning Technologies
Dates: 6-7 July, 2020
Location: Online Conference
Abstract:
Developing social learning skills is one of the basic aims of the contemporary school system. The process of mastering such skills by students is time consuming for teachers. Video recording of the students' activities followed by the analysis of the video and the corresponding future planning the skills development can reduce the time of learning thereby increasing its effectiveness. However, in the personalized learning process, such a straightforward video recording has to comprise recording of each of the students, which is almost impossible.

The paper proposes a solution of the above problem. The solution uses an intelligent machine learning tool. The proposed tool functions on the base of intelligent processing of the 3600 video recording of students’ cooperation activities. Such processing enables to detect each student's video from the global video stream and to process it separately. It allows to recognize patterns of behavior that the student demonstrates during the exercise. The intelligent tool find and highlights representative patterns that are eligible and/or ineligible according to the role of the student in the activity. The intelligent video processing allows specifying sets of specific actions that each member of the group demonstrate according to his/her roles in the team. After finishing the activity, each student immediately get personal feedback that reflects his/her success and failures during the activity.

The proposed deep learning based approach of the pattern recognition was implemented as a software application. This application uses a quite advanced hardware and software resources that enables a corresponding recording and processing. Nevertheless, students use a conventional equipment (smartphone, laptop) to get the feedback.

The preliminary experimental results of our study are positive. They demonstrate encouraging prospects of integration intelligent tools to support effective reflection of the students’ teamwork in elementary school.
Keywords:
Intelligent tool, machine learning, soft skills, social skills, teamwork.