ISSN 2348-1196 (print) International Journal of Computer Science and Information Technology Research ISSN 2348-120X (online) Vol. 10, Issue 3, pp: (26-33), Month: July - September 2022, Available at: www.researchpublish.com
Cyber Attack detection Using Big data analysis Yazeed Al Moaiad1, Yasser Mohamed Abdelrahman Tarshany2, Nasir Ahmed Algeelani3, Wafa Al-Haithami4 Faculty of computer science and Information technology, Faculty of Islamic Sciences, AL-Madinah International University, Kuala Lumpur, Malaysia DOI: https://doi.org/10.5281/zenodo.6924399
Published Date: 28-July-2022
Abstract: Network-based Intrusion Detection System is a threat caused by the explosion of computer networks and the myriad of recent content-based threats, which occur daily. As well as an overview of machine learning approaches for signature and anomaly detection methods, this article discusses several machine learning strategies applied to intrusion detection and preprocessing. The NIDS taxonomy and attribute classifier have created classifications and outlines. Machine learning methods are widely utilized in anomaly detection using many data sets. Additional preprocessing methods have been added, for example, sorting and discretization have been applied to the data collection of measured values. Custom methods focused on search algorithms using machine learning that uses novel search algorithms are vulnerable to being revealed. This analysis is highly relevant to the use of machine learning methods used in computer security, which furthers their cause. Keywords: Intrusion Detection, K-means Algorithm, Machine learning, Swarm Intelligence.
I. INTRODUCTION If the number of network providers grows, their stability, integrity, and availability often become an issue, which is an increasingly important concern. The 2014 Cisco Security Report (CISCO, 2014), where the increase in weaknesses is noted, that taking advantage of the latest assault approaches and revamped tactics has resulted in a corresponding rise in security threats, backs up this information. Lastly, the report suggests that organizations are no longer able to protect their networks. In comparison, 100% of the world's networks were found to have harmful material on their web servers, and 96% of the traced networks were found to be “phishing”. Distributed denial of service (DDoS) attacks that target websites or the Internet have been even more common in 2013 and have risen in both frequencies and sophistication. Disorganized assaults have given way to more focused activities by cybercriminals, so complex that they might undermine the national security of both private and public institutions, as well as the country's prestige. Additionally, there is an increase in the vulnerability and response footprint as a result of the rapid increase in connected devices and virtualized Cloud computing environments. Vulnerabilities and security technologies have been supplemented by various kinds of mobile devices and infrastructures, making attacks by previously unseen adversaries possible. As cybercriminals have found, the Internet's technology offers much more advantages than only looking for individual targets. These infrastructure attacks attempt to access the main web, DNS, and data center servers to potentially propagate risks to numerous smaller properties that are dependent on them. Warming up diminishes faith in the services which support the Internet [5]. It's perpetrated by trespassers. There are two groups of attackers: those that target unauthenticated computer networks, and others that have authenticated access to authenticated ones. For this purpose, thus, a shield is needed to defend the device from intruders [14]. Any effort to attack the resources, such as secrecy or availability, and honesty that undermines or
Page | 26 Research Publish Journals