Detecting Cyber Attacks by Applying MachineLearning Techniques

Abstract

Because of the growth in cloud services, the growing number of web applications customers, and modifications to network infrastructure that connects devices running several operating systems, cyber security is facing new challenges. For this reason, network security components, sensors and insurance conspiracies must also be developed to meet the needs and concerns of the customers in order to combat new threats. In this post, we focus on combating application layer cyber assaults, which are rankedas the most dangerous threats and the most important test for network and cyber security. Most of this essay focuses on machine learning as a method to cope with model normal use and to detect cyber threats. Chart-based division technique and dynamic programming were used in order to obtain examples in the form of Perl Compatible Regular Expressions (PCRE) ordinary expressions. Information is gathered through HTTP requests made by the client to a web worker, which is used in the model. Using the CSIC 2010 HTTP Dataset, we've been able to demonstrate the effectiveness of our approach.