IJFANS International Journal of Food and Nutritional Sciences

ISSN PRINT 2319 1775 Online 2320-7876

REVIEW ON CYBER NETWORK THREAT DETECTION SYSTEMS USING DIFFERENT TYPES OF ADVANCED TECHNIQUES

Main Article Content

RadhaRani Akula, GS Naveen Kumar

Abstract

The establishment of an automated and effective cyber-threats detection technique is one of the major challenges in cyber safety. In this research, we describe a novel approach to finding cyberthreats that is based on artificial semantic networks. The suggested method uses a deep learning-based detection strategy for accelerated cyber-threat discovery and converts a wide variety of accumulated security events into private event accounts. This research study work's goal is to provide an analysis of some of the frequently used device learning algorithms used to spot some of the most dangerous cyberthreats to the internet. Deep belief networks, decision trees, and support vector machines—three fundamental AI techniques—are typically under investigation. In order to identify an invasion or assault, a Breach Discovery System (IDS) examines the link records and traffic control packets. The volume of records produced by a network is enormous. The IDS extracts attributes from the records and then categorises them to determine if the record or connection is part of an attack or regular online traffic. It is feasible to reduce the attribute's dimension to aid in the equipment learning approaches used for category. Establishing general and systematic strategies for classifying breach finding is offered in this research study. The key recommendations are to employ information mining techniques to uncover recurring patterns in the system characteristics that define network behaviour as well as to use the set of relevant system features to spot anomalies and known intrusions.

Article Details