Suggested Cyber-Security Strategy That Maximizes Automated Detection of Internet of Things Attacks Using Machine Learning

Suggested Cyber-Security Strategy That Maximizes Automated Detection of Internet of Things Attacks Using Machine Learning

Dharmesh Dhabliya, Pratik Pandey, Varsha Agarwal, N. Gobi, Anishkumar Dhablia, Jambi Ratna Raja Kumar, Ankur Gupta, Sabyasachi Pramanik
Copyright: © 2024 |Pages: 14
DOI: 10.4018/979-8-3693-1062-5.ch010
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

The world is experiencing an unparalleled digital revolution because of the advancement of computer systems and the internet. This change is made even more noticeable by the fact that the internet of things is opening up new business options. However, the rise of cyberattacks has severely harmed system and data security. It is true that computer intrusion detection systems are automatically activated. However, due to its conceptual flaws, the security chain is insufficient to counter such attacks. It prevents the full potential of machine learning from being realized. Therefore, a new framework is required to properly safeguard the IT environment. The goal in this regard is to use machine learning methods to build and execute a new strategy for cyber-security. The goal is to improve and maximize the identification of harmful assaults and intrusions in the internet of things. Following the application of this novel strategy on the Weka platform, the authors get a final model that is reviewed and evaluated for performance.
Chapter Preview
Top

As part of its plan to enhance living circumstances and provide individuals with disabilities, particularly those with autism, with access to educational programs, Mauritania has established care facilities to lessen their suffering and enhance their educational opportunities. For instance, the Nouakchott autism spectrum disorder facility is home to 66 children, eight of them are females. The children are split into two levels: beginner and intermediate, and they are overseen by fifteen specialized educators and one expert. Data is gathered and taught utilizing supervised, semi-supervised, and unsupervised learning techniques to create intrusion detection systems (Chua & Salam, 2022). The long-term performance assessment of IDS is suggested in this article. The objective is to be able to identify zero-day attacks that are not yet known.

However, machine learning techniques are used in a synthesis on the examination of Cloud Computing dangers, issues, and security solutions (Butt et al., 2020). They are used in hardened, semi-supervised, supervised, and unsupervised modes to address cloud security concerns.

Numerous studies have been conducted on the industrial Internet of connected objects, or I-IoT, which is also an active area of study. Thus, this research addresses the issue of poor detection rates and large false alarm proportions (Khan et al., 2022). This work's only goal is to identify and thwart cyberattacks. The primary emphasis is not on issues pertaining to the costs and effects of this discovery.

A significant contribution to the solution of the linked objects security challenge is made in (William et al, 2022). They are acknowledged with an exhaustive review of the literature. There is little doubt that the goals and intentions of the studies included in this analysis vary. Some of them address the issue from the logical perspective of the inherent technological limitations of the Internet of Things, namely with regard to energy, memory, and storage.

Complete Chapter List

Search this Book:
Reset