Early Dropout Prediction in MOOCs through Supervised Learning and Hyperparameter Optimization

Theodor Panagiotakopoulos,Sotiris Kotsiantis,Achilles Kameas,Georgios Kostopoulos,Omiros Iatrellis

doi:10.3390/electronics10141701

Theodor Panagiotakopoulos, Sotiris Kotsiantis + Show 3 more

Open Access

PDF Available

https://doi.org/10.3390/electronics10141701

Copy DOI

Export

Save

Cite

Abstract
Highlights/Summary
Full-Text PDF
Similar Papers

Abstract

Listen

Over recent years, massive open online courses (MOOCs) have gained increasing popularity in the field of online education. Students with different needs and learning specificities are able to attend a wide range of specialized online courses offered by universities and educational institutions. As a result, large amounts of data regarding students’ demographic characteristics, activity patterns, and learning performances are generated and stored in institutional repositories on a daily basis. Unfortunately, a key issue in MOOCs is low completion rates, which directly affect student success. Therefore, it is of utmost importance for educational institutions and faculty members to find more effective practices and reduce non-completer ratios. In this context, the main purpose of the present study is to employ a plethora of state-of-the-art supervised machine learning algorithms for predicting student dropout in a MOOC for smart city professionals at an early stage. The experimental results show that accuracy exceeds 96% based on data collected during the first week of the course, thus enabling effective intervention strategies and support actions.

Highlights

Over recent years, massive open online courses (MOOCs) have gained increasing popularity in the field of online education
Despite the potential benefits of MOOCs, these courses are characterized by low retention rates, which directly affect student success [3]
Hyperparameter optimization was automatically performed on a set of optimal hyperparameters for all the examined learning algorithms using random search, which chooses random groupings of hyperparameters for training the learning model

Summary

Introduction

Massive open online courses (MOOCs) have gained increasing popularity in the field of online education. Students with different needs and learning specificities are able to attend flexible and high-quality online courses offered by universities and educational institutions [1]. These courses vary significantly from the traditional online courses delivered by higher education institutions, in terms length and content structure [2]. Despite the potential benefits of MOOCs, these courses are characterized by low retention rates, which directly affect student success [3]. A number of studies have highlighted various factors influencing retention rates in MOOCs. Student motivation, challenge, future economic profit, growth of personal and professional identity, insufficient background knowledge, and lack of time have a significant impact on preventing students from completing a MOOC. Lack of mobile-friendly features (e.g., inability to watch videos via smart phones), lack of interaction with peers and instructors, and difficulty in following the language of the instructor have a major influence on student dropout [4]

Objectives

Results

Conclusion