The Use of Students’ Digital Portraits in Creating Smart Higher Education: A Case Study of the AI Benefits in Analyzing Educational and Social Media Data

Abstract

The problem of a smooth and synchronous integration of smart education into the traditional educational system of higher education institutions (HEIs) and transforming it into a person-centered, blended, systematic adaptive educational environment is becoming a new paradigm of higher education institutions (HEIs). But there are a few studies that discuss the role of artificial intelligence (AI) in providing comprehensive and predictive analysis not limited to educational outcomes but focused on students’ individual portraits and aimed at improving the quality of education and provision of an integrated educational environment. This paper investigates role, application, and challenges of applying AI to targets and features of smart education concept: proactive, predictive analysis based on instruments aimed at building up a system that not only records students’ performance but also predicts and influences it. This study offers the concept of the comprehensive system of the university targeting to provide predictive analysis of students’ performance based on quantitative and qualitative information from multiple sources concerning students’ education, research performance, and social network information joined into unified, comprehensive dataset of each student (digital portraits) that allows to design individual tracks and use advising system to correct this track according to students’ needs and personal goals. As a result, we can predict students’ dropout with almost 90% accuracy. The main research problem of the study is to observe the implementation of AI instruments in higher education according to the needs and goals of universities and personal goals of each student and investigate the impact of this in blending different types of educational activity and creation of the integrated educational environment. Our findings are based on the analysis of the information about more than 20,000 students in 2013–2019.