An Intelligent Risk Taxonomy for Enterprise Cybersecurity Using Advanced Deep Learning Techniques

The exponential growth of advanced cyber-attacks and constraints on the traditional approach has made it necessary for enterprises to employ intelligent and adaptive approaches in their cyber security operations. The current paper introduces An Intelligent Risk Taxonomy for Enterprise Cyber security Using Advanced Deep Learning Techniques for effective detection. The input data is first gathered from the IDS Intrusion Dataset, and it is then fed into the reprocessing section to eliminate any missing information and sanitize the data. Enhanced Zebra Gooseneck Barnacle Optimization is used to choose the ten pertinent features from the pre-processed data. The features are fed to the classification segment by using Scalable and Adaptive Graph Neural Network (SAGNN), in this segment predict the Risk Taxonomy for Enterprise Cybersecurity and classify it DoS, DDOS, Brute Force, SQL Injection, Infilteration, Benign and Bot. The proposed method is implemted in PYTHON and compare to other exiting methods. The proposed method attains 98.3 accuracy and lower error rate.