Khor, Ean Teng

The current study set out to understand the factors that explain working adults’ microlearning usage intentions using the Decomposed Theory of Planned Behaviour (DTPB). Specifically, the authors were interested in differences, if any, in the factors that explained microlearning acceptance across gender, age and proficiency in technology. 628 working adults gave their responses to a 46-item, self-rated, 5-point Likert scale developed to measure 12 constructs of the DTPB model. Results of this study revealed that a 12-factor model was valid in explaining microlearning usage intentions of all working adults, regardless of demographic differences. Tests for measurement invariance showed support for invariance in model structure (configural invariance), factor loadings (metric invariance), item intercepts (scalar invariance), and item residuals (strict invariance) between males and females, between working adults below 40 years and above 40 years, and between working adults with lower technology proficiency and higher technology proficiency levels. While measurement invariance existed in the data, structural invariance was only found across gender, not age and technology proficiency. We then assessed latent mean differences and structural path differences across groups. Our findings suggest that a tailored approach to encourage the use of microlearning is needed to suit different demographics of working adults. The current study discusses the implications of the findings on the use and adoption of microlearning and proposes future research possibilities.

The COVID-19 pandemic induced a digital transformation of education and inspired both instructors and learners to adopt and leverage technology for learning. This led to online learning becoming an important component of the new normal, with home-based virtual learning an essential aspect for learners on various levels. This, in turn, has caused learners of varying levels to interact more frequently with virtual resources to supplement their learning. Even though virtual learning environments provide basic resources to help monitor the learners’ online behaviour, there is room for more insights to be derived concerning individual learner performance. In this study, we propose a framework for visualising learners’ online behaviour and use the data obtained to predict whether the learners would clear a course. We explored a variety of binary classifiers from which we achieved an overall accuracy of 80%–85%, thereby indicating the effectiveness of our approach and that learners’ online behaviour had a significant effect on their academic performance. Further analysis showed that common patterns of behaviour among learners and/or anomalies in online behaviour could cause incorrect interpretations of a learner’s performance, which gave us a better understanding of how our approach could be modified in the future.

Purpose The purpose of the study is to build predictive models for early detection of low-performing students and examine the factors that influence massive open online courses students' performance. Design/methodology/approach For the first step, the author performed exploratory data analysis to analyze the dataset. The process was then followed by data pre-processing and feature engineering (Step 2). Next, the author conducted data modelling and prediction (Step 3). Finally, the performance of the developed models was evaluated (Step 4). Findings The paper found that the decision trees algorithm outperformed other machine learning algorithms. The study also confirms the significant effect of the academic background and virtual learning environment (VLE) interactions feature categories to academic performance. The accuracy enhancement is 17.66% for decision trees classifier, 3.49% for logistic regression classifier and 4.89% for neural networks classifier. Based on the results of CorrelationAttributeEval technique with the use of a ranker search method, the author found that the assessment_score and sum_click features are more important among academic background and VLE interactions feature categories for the classification analysis in predicting students' academic performance.

Predictive modelling in the education domain can be utilised to significantly improve teaching and learning experiences. Massive Open Online Courses (MOOCs) generate a large volume of data that can be exploited to predict and evaluate student performance based on various factors. This paper has two broad aims. Firstly, to develop and tune several Machine Learning (ML) models to perform classification tasks on the dataset to predict student performance, including Linear Regression, Logistic Regression, Random Forests, K-Nearest Neighbours, and more. Secondly, to evaluate the efficacy of these ML models and identify those which are best suited to this task. The categories of data utilised in achieving these aims include (i) demographic information, (ii) academic background, and (iii) interaction with MOOC course materials. The research procedure comprises five phases: data exploration to analyse the dataset, feature engineering which involves discerning the most important features and converting them into a format decipherable by the ML models, model building, model evaluation by measurement of accuracy, and subsequent comparative evaluation between the different models. The results achieved in this study are expected to have implications on how MOOC platforms utilise data to improve user experience. As indicated by the findings of this study, the data collected by these platforms may be used to predict performance with accuracy of over 77%; this extracted information can be exploited to enhance educational theory or practices in the context of MOOCs, for instance by implementing varying teaching methodologies or providing different types of resources based on predicted performance.

Data mining is gaining increasing traction in the field of education as its applications in the education sector has increased over the past few years. Different data mining methods can be used to gain insights into educational data, including the uncovering of hidden patterns and prediction of output. The methods include classification analysis, association rule learning, anomaly or outlier detection, clustering analysis, and regression analysis. In this study, the classification analysis is used with decision tree algorithms to predict learners' performance. The findings reveal that the algorithm can be used to build a predictive model with good performance measure based on accuracy level, true positive (TP) rate, and false positive (FP) rate.

In personalized learning (PL), learning processes are customized to account for student skills and preferences. However, as PL is generally based on a single data type, it cannot wholly represent students' learning behaviors and progress. Hence, it is crucial to leverage Multimodal Learning Analytics (MMLA) in PL to alleviate these restrictions. A systematic literature review was conducted to explore the use of MMLA in PL and investigate its benefits across several contexts and approaches. The underexplored aspects of MMLA in PL, like the gaps in topics, pedagogies, learning settings and environments, populations, and modalities studied, are addressed, and MMLA’s potential to provide real-time tailored feedback and improve engagement is discussed.

Learning analytics methods are widely applied in the educational field to gain insights on hidden patterns from educational data. Methods like predictive learning analytics are used to identify and measure patterns in learning data and extrapolate future behaviours. It can be used to enable the learners to be more self-aware of their learning behaviours and to enable the instructor to take appropriate actions informed by the trace of data. Thus such methods can empower learners as they progress through online training, and allows them to be self-regulated in order to solidify their learning and develop positive habits that will enhance their learning experiences. This paper reports on the use of a popular decision tree classification algorithm using behavioural features from a public domain dataset to develop a predictive model for predicting learning performance. Among the five behavioural features, we find that the measure of visited resources provides the most discriminating rules in the classifier.

The study aims to gain insights into the patterns related to the diversity of learners and evaluate the relationship between learners’ factors and their academic performance. The pattern discovery was performed by applying clustering data mining to obtain typologies of the learners based on the academic records and e-learning interaction behaviour feature category. In this study, the k-means clustering unsupervised machine learning algorithm was applied to obtain the clusters. The clustering analysis of learners’ academic records and interaction behavioural patterns between learner sub-populations allows for a better understanding of how the learners behave and achieve. The clustering results identified similar group learners, and the learners could be provided with appropriate educational supports and approaches to enhance learners’ learning experience. The findings of this study are also useful to understand the effects of different features on learners’ academic performance specifically in an e-learning environment.

Machine learning is gaining increasing popularity in the education sector due to its potential to improve various aspects of the education system. The present study aims to develop eight types of predictive models to forecast students’ academic performance. The first four predictive models were developed based on the (i) demographical features, (ii) academic background features, (iii) parents support features, and (iv) learning behaviour features and the other four predictive models were generated based on the (i) demographical features, (ii) academic background features, and (iii) parents support features. The machine learning algorithms used to develop the models include the decision tree, the naïve Bayes, the support vector machine and the neural network. From the result analysis of the dataset, it showed that support vector machine algorithm performs high prediction on students’ academic performance. The findings also revealed that learning behaviour features are significant and have a greater impact on students’ academic performance.

The paper presents a data mining approach using unsupervised learning for profiling students. Unsupervised learning specifically the K-means clustering algorithm is applied to obtain clusters with similar patterns and characteristics. The clustering experiments were performed using academic background, parental support, and learning behavioural features as attributes. The characteristics that distinguish students belonging to those different clusters were examined. The findings uncovered the key characteristics of students’ performance, and it is helpful for future prediction. Appropriate learning support and intervention could be provided to tailor to the individual cluster of students to enhance their performance. The clustering algorithm also serves as a potential benchmark to monitor the progress of students’ performance and helps teachers to improve the course success.