7 September 2021 in IEEE Access
The Internet of Things (IoT) has rapidly transformed digital environments across a multitude of domains with increased connectivity and pervasive virtualization. The distributed computing paradigm of Edge Computing has been postulated to overcome the concerns of response time, bandwidth, energy consumption, and cybersecurity. In comparison to the other concerns, limited studies have focused on cybersecurity, mainly due to the inherent complexity of threat detection at the Edge. However, the widespread adoption of IoT applications in economic, social, and political contexts is a stringent indication of the significant impact from cyber-attacks. This paper aims to address this challenge by presenting an effective and efficient machine learning approach for threat detection at the Edge of IoT. The novel contributions of this approach are, a new Enhanced Geometric Synthetic Minority Oversampling Technique (EG-SMOTE) algorithm to resolve the imbalanced distribution of data streams at the IoT Edge, an extension to the Growing Self Organizing Map (GSOM) algorithm based on Hyperdimensional Computing for energy efficient machine learning from unlabeled data streams. The proposed EG-SMOTE + GSOM approach has been tested using four open access datasets; three benchmark, KDD99 (F-Score = 0.9360), NSL-KDD (F-Score = 0.9647), CICIDS2017 (F-Score = 0.9999), and one industry-focused botnet IoT traffic dataset, BoT-IoT (F-Score = 0.9445). The EG-SMOTE approach has outperformed SMOTE and G-SMOTE approaches in a vast number of experiments that are tried with different classifiers. The results of these experiments confirm the novelty, efficiency and effectiveness of this approach for cybersecurity at the IoT Edge.
17 October 2025 in European Conference on e-Learning
The key challenges in the delivery of distance learning programs include maintaining student engagement and ensuring that students achieve intended learning outcomes. While video content has the advantage over lecture notes in being able to hold a student’s attention more effectively and improve comprehension, the preparation of high-quality pre-recorded content can involve a large initial time investment, while infrastructure constraints such as availability of devices and network connectivity may also hold, especially in developing countries. This research studies the transition of two course modules offered in the first year of a distance learning program in Sri Lanka from a 100% notes-based delivery style to the incorporation of pre-recorded videos for each of the topics in the modules. Specifically, the study investigates the impact of the inclusion of video content into the course content on learning behavior measured by interactions with the LMS and learning outcomes measured by continuous assessment and written examination marks. The study uses data from 12 cohorts of students and two modules, where the first five cohorts did not have access to any videos, while the remaining cohorts did. The impact of the availability of video content is assessed using a repeated-cohort multiple regression model where changes to the examiner of the module can be controlled for. Results of this study show that marks for non-formative quizzes and summative continuous assessments are positively and significantly associated with the minutes of video content in both modules. However, the impact on end-semester examinations and submission of formative assignments varies across modules and is even negative for one subject. These findings suggest that the incorporation of video content perhaps has a more substantial short-term effect on learning outcomes, helping with immediate retention of content, though the evidence about longer-term effects, as observed in final examinations, is mixed. While further work is needed to incorporate additional course modules and perhaps use detailed topic-wise marks from final examinations, the findings of this research provide a case study for the use of data-driven strategies for content design and development.
1 June 2026 in EDULEARN Proceedings
1 June 2026 in EDULEARN Proceedings
1 June 2025 in EDULEARN Proceedings