Enhancing Network Security using Possibility Neutrosophic Hypersoft Set for Cyberattack Detection

Network security is any endeavor intended to defend the integrity and usability of the data and network. Fast development in network technology and the scope and amount of information transported on a network is gradually growing. Based on these situations, the complexity and density of cyber-attacks and threats are also increasing. The constantly expanding connectivity makes it more difficult for cyber-security specialists to monitor all the movements on the network. More complex and frequent cyber-attack makes anomaly identification and detection in network events challenging. Machine learning (ML) provides different techniques and tools to automate cyber-attack detection and for prompt prognosis and analysis of attack types. The model of a neutrosophic hypersoft set (NHSS) is a combination of a neutrosophic set with a hypersoft set. It is a useful structure to handle multi-objective problems and multi-attributes with disjoint attributable values. This study derives the Possibility Neutrosophic Hypersoft Set for Cyberattack Detection (pNHSS-CAD) technique to improve network security. The pNHSS-CAD method has its formation in feature selection with the Whale Optimization Algorithm (WOA), which successfully recognizes the important features from the data, thus improving processing speed and reducing dimensionality. Following feature selection, the pNHs-set classifier is employed for the robust detection and identification of cyber-attacks, which leverages the power of the neutrosophic set to deal with ambiguity and uncertainty in the information. The Firefly (FF) technique is applied for hyperparameter fine-tuning, which ensures the model operates at maximum effectiveness to enhance the performance of the classification. This wide-ranging method leads to a very efficient cyberattack recognition method, which can able to accurately mitigate and identify risks in the real world