Text analytics on MOOCs. A comprehensive analysis of emotions

Abstract

The value of diversity in education is highly emphasized in recent years, particularly in the wake of the COVID-19 pandemic, by many scholars. Massive open online courses (MOOCs) have aided the evolution of online learning by broadening the range of learning opportunities available. They have gained popularity, especially in higher education by providing unlimited access to lectures and rich learning materials by renowned and respected academics in a wide variety of areas, with no restrictions and at very low fees. Furthermore, learners' motivations for enrolling in a MOOC may vary depending on their choices for the course's instructional design as well as their emotions. Knowing this, the development of more effective online courses that address affective concerns would appeal to a wider audience and improve the learning experience. This research aims to uncover the emotional characteristics of MOOCs to better understand why learners choose a specific course among hundreds of options available on MOOC sites. For extracting the learners' emotions from user reviews, the study used Kansei Engineering approach, which is enhanced with text analytics techniques. The research methodology entails gathering reviews from MOOCs and analyzing them using natural language processing (NLP) techniques to discover Kansei words that characterize MOOCs, notably for courses in the discipline of Data Science. The expected output of this study is a Kansei corpus for online courses in this discipline.