Learners Demographics Classification on MOOCs During the COVID-19: Author Profiling via Deep Learning Based on Semantic and Syntactic Representations.

Tahani Aljohani,Alexandra I Cristea

doi:10.3389/frma.2021.673928

Tahani Aljohani, Alexandra I Cristea

Open Access

https://doi.org/10.3389/frma.2021.673928

Copy DOI

Abstract

Massive Open Online Courses (MOOCs) have become universal learning resources, and the COVID-19 pandemic is rendering these platforms even more necessary. In this paper, we seek to improve Learner Profiling (LP), i.e. estimating the demographic characteristics of learners in MOOC platforms. We have focused on examining models which show promise elsewhere, but were never examined in the LP area (deep learning models) based on effective textual representations. As LP characteristics, we predict here the employment status of learners. We compare sequential and parallel ensemble deep learning architectures based on Convolutional Neural Networks and Recurrent Neural Networks, obtaining an average high accuracy of 96.3% for our best method. Next, we predict the gender of learners based on syntactic knowledge from the text. We compare different tree-structured Long-Short-Term Memory models (as state-of-the-art candidates) and provide our novel version of a Bi-directional composition function for existing architectures. In addition, we evaluate 18 different combinations of word-level encoding and sentence-level encoding functions. Based on these results, we show that our Bi-directional model outperforms all other models and the highest accuracy result among our models is the one based on the combination of FeedForward Neural Network and the Stack-augmented Parser-Interpreter Neural Network (82.60% prediction accuracy). We argue that our prediction models recommended for both demographics characteristics examined in this study can achieve high accuracy. This is additionally also the first time a sound methodological approach toward improving accuracy for learner demographics classification on MOOCs was proposed.

Highlights

The wave of technological innovations has affected education systems, with one output being the socalled Massive Open Online Courses (MOOCs)
Our main research questions are: 1-How can deep learning methods be designed to predict the demographic characteristics of learners in a MOOC, based on the comments they exchange in the discussion forum? This has been done by ultising Ensemble learning of CNN and RNN for employment profiling, and Recursive NN for gender profiling, with both models fed by textual features extracted from comments. 2-What are the most important demographics in MOOCs that are needed for almost all MOOC researches? We discussed in our investigation about the importance of employment and gender as demographics variables for MOOC researches and studies
We tackle here the difficult problem of predicting the gender of learners based on their comments only which are often available across MOOCs

Summary

Introduction

The wave of technological innovations has affected education systems, with one output being the socalled Massive Open Online Courses (MOOCs) They are educational information systems providing a way to democratize knowledge, by usually providing free learning, which successfully attracts significant numbers of users. Owing to this phenomenon, users in MOOCs are very varied in terms of age, gender, employment status, level of education, etc. We investigate learners’ posts from a different angle than other works in MOOCs. Our research target is the heterogeneousness of MOOC environments, in terms of their learner demographics based on employment status and gender. Course content could be personalised based on gender differences, since these differences have been already proven to be one of factors for completing a course in MOOCs, but there are differences which depend on the course (Morris et al 2015)

Objectives

Results

Conclusion