Analysis of Factors Influencing the Outcomes of AR Based Education Approach in STEM Learning Using Machine Learning

Packirisamy, Priyadharsini (2024) Analysis of Factors Influencing the Outcomes of AR Based Education Approach in STEM Learning Using Machine Learning. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (3MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

Augmented reality (AR) based education approach is an evolving domain. Several studies have been conducted to analyse the impacts of AR-based education. However, these studies have certain limitations, like usage of rudimentary techniques such as basic mathematical and statistical methods and analysing the generic binary outcomes like positive or negative impacts. In practice, to implement a successful AR-based education, numerous factors will weigh in. Few examples are knowledge of technology, accessibility to the application, subject matter interest, and infrastructure challenges. Another limitation observed in these studies are the lack of adequate data both in terms of number of participants and data dimensionality. The aim of this research is to identify the potential factors that influence the outcomes of AR-based education approach which will facilitate addressing the areas of improvement. Machine learning algorithms have been used to determine these factors, as they are well suited to identify the patterns and to solve real-world problems effectively. This overcomes the inadequacy of the basic analysis techniques that were previously used. For fair analysis, an extensive dataset collected as part of the ARETE’s Pilot 2 project with 1988 participants across 11 countries and varied age groups for two subjects - math and science, divided as intervention and control groups have been chosen. This fills the gap in previous studies of inadequate data. The data has been preprocessed, and existing trends in the data has been analysed using exploratory data analysis. This data has been used to train the neural networks RNN, CNN, and LSTM and the machine learning model SVM to identify the influencing factors. The overall accuracy yielded by the models are 99%, 98%, 96%, and 94% for RNN, CNN, LSTM, and SVM respectively. This research revealed that age, gender, test book of the subject, media usage and availability of technology are some of the most influential factors in the AR-based learning approach. Also, the overall impact shows positive attitude towards the AR-based education approach.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Rustam, Furqan UNSPECIFIED
Subjects:	L Education > L Education (General) Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning G Geography. Anthropology. Recreation > GV Recreation Leisure > Games and Amusements > Online Games
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	25 Aug 2025 08:45
Last Modified:	25 Aug 2025 08:45
URI:	https://norma.ncirl.ie/id/eprint/8603

Actions (login required)

View Item