DEEP LEARNING APPROACHES FOR IDENTIFYING CYBER ATTACKS IN DATA SCIENCE
S. Raja Ratna, G. Gangadevi, J. Jospin Jeya · 2023
Cyber-attacks are a growing hazard to cyber- physical systems, and firms that handle sensitive data must be able to identify and anticipate these attacks. Because the types of Cyber Attacks are becoming more diverse, anti-virus scanners alone are no longer adequate to offer protection. The growth of Internet of Things (IoT) devices has led to significant increases in cyber security vulnerabilities, and the market for software is growing as more software is used in more aspects of daily life. Hackers launch cyber-attacks in a variety of ways, including Phishing, Dos, R2L, Probing, Malware, and U2R. Large data sets can occasionally be compromised without the affected parties' knowledge, which can cause serious interruptions to business continuity and large financial losses. By examining security data, Machine Learning technology is essential for providing an automated, dynamically improved, and up to security system. Large datasets may be handled by Machine Learning technology, which also provides greater visualization capabilities. Due to built-in weaknesses and threats that can be used at any stage in the system, supply chain security is difficult. Our study's findings demonstrate that the majority of rules exhibit steady support and confidence values, enabling the prediction of Cyber Attacks over a period of days without the need for daily rule updates.