Work

Analyzing Student Behavioral Patterns in MOOCs Using Hidden Markov Models in Distance Education

Year: 2024

Type: article

Abstract: The log files of Massive Open Online Courses (MOOCs) reveal useful information that can help interpret student behavior. In this study, we focus on student performance based on their access to course resources and the grades they achieve. We define states as the Moodle resources and quiz grades for each student ID, considering participation in resources such as wikis and forums. We use efficient Hidden Markov Models to interpret the abundance of information provided in the Moodle log files. The transitions among certain resources for each student or groups of students are determined as behaviors. Other studies employ Machine Learning and Pattern Classification algorithms to recognize these behaviors. As an example, we visualize these transitions for individual learners. Additionally, we have created row and column charts to present our findings in a comprehensible manner. For implementing the proposed methodology, we use the R programming language. The dataset that we use was obtained from Kaggle and pertains to a MOOC of 4037 students.

Source: Applied Sciences

DOI: 10.3390/app142412067

Authors: Vassilios S. Verykios, Nikolaos S. Alachiotis, Evgenia Paxinou, Georgios Feretzakis

Institution: Hellenic Open University

Language: English

Topic: Online Learning and Analytics

Subfield: Computer Science Applications

Field: Computer Science

Domain: Physical Sciences

SDG: Quality education

Open Access status: gold

Estimated APC paid: $2,728