Work

Securing the IoT: A Machine Learning Approach to Cyber Threat Analysis

Year: 2024
Type: book-chapter
Abstract: The rise of Internet of Things (IoT) devices [1] has brought about numerous conveniences, but it has also introduced security challenges, notably the alarming threat of botnets.These malicious networks can compromise a large number of devices, posing significant risks to privacy, data integrity, and network availability.Recent years have seen the effectiveness of Machine Learning (ML) techniques in addressing this concern by identifying and mitigating IoT botnet attacks.The proposed system utilizes both supervised and unsupervised ML algorithms, such as classification, decision trees, and logistic regression, to identify compromised IoT devices.Trained on carefully labelled datasets, the system learns distinctive features and patterns associated with malicious activities, employing feature engineering to enhance accuracy.Real-time monitoring and anomaly detection are integrated to promptly respond to botnet-related activities.Ensemble learning methods further strengthen the system, resulting in high accuracy and minimized false alarms, contributing significantly to the security of IoT networks and paving the way for future advancements in IoT security.
Language: English
Open Access status: hybrid
OpenAlex ID: W4401071752
FWCI:
Cites: 9
Cited by:
Related to: 10