Copy the page URI to the clipboard
Bandara, Indra; Bowkis, Michael and Ioras, Florin
(2024).
DOI: https://doi.org/10.21125/edulearn.2024.2165
Abstract
In the current digital era, the ability to differentiate between authentic and malicious content is important, particularly with the rapid spread of information via social media. In the field of cybersecurity, the threat posed by malicious websites, which engage in activities like spam, phishing, and drive-by downloads, is a real concern. These websites can lead to various negative outcomes, including financial loss, privacy breaches, and malware infections, and have an economic impact. In this study, the authors examine the use of machine learning (ML) and deep learning (DL) for the detection of such websites. We focus on Long Short-Term Memory (LSTM) networks, a type of recurrent neural network, for their capability to identify and address malicious web activities. The study considers various ML and DL approaches, providing insights and recommendations for developing effective malicious website detection systems. Additionally, the authors look at how these technologies can benefit online education by ensuring the authenticity of digital content. Advanced detection models can help educators and online platforms filter out false content, aiming to provide accurate and trustworthy educational resources. The study highlights the relevance of addressing malicious web activities and suggests that LSTM and other deep learning techniques could improve information credibility in our interconnected world.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)
