Cloud Worm Detection and Response Technique By Integrating The Enhanced Genetic Algorithm An Threat Level
A worm is known as a malicious code that can replicate, infect and propagate itself without attaching itself to any host. Unlike other malicious codes, such as a Trojan horse and a virus, a worm can cause serious damage due to its payload and interrupt or stop cloud resources and services which then...
Saved in:
| Summary: | A worm is known as a malicious code that can replicate, infect and propagate itself without attaching itself to any host. Unlike other malicious codes, such as a Trojan horse and a virus, a worm can cause serious damage due to its payload and interrupt or stop cloud resources and services which then lead to loss of money, confidential information and productivity for organisations or users that rely on data storage, services and applications that run on the cloud. Detecting and stopping cloud worm attacks have become a hard and challenging endeavour. Therefore, this research proposes a cloud worm classification based on its features. The research also develops a cloud worm detection technique by integrating the enhanced genetic algorithm and proposes a cloud response technique based on threat levels. The proposed cloud worm detection and response technique is evaluated based on accuracy rates. The novelty and strengths of the proposed technique for cloud worm detection and response lies in the worm cloud classification, the integration of the enhancement genetic algorithm and the threat level measurement that impact confidentiality, integrity and availability. For the enhanced genetic algorithm (EGA), new parameters have been constructed for the existing genetic algorithm based on the selection proportional of fitness, tree mutation tree crossover and evolution controller to increase the 's accuracy rate of the cloud worm detection technique. The genetic algorithm is based on the idea of the natural selection process and genetics which imitates the biological process, thereby
helping solve unconstrained and constrained optimisation problems. The method of new offspring creation from the fittest parents has been mapped and used to detect unknown future cloud worm attacks. Threat level is measured by security metrics on the basis of confidentiality, integrity and availability impact. An experiment has been conducted in a controlled lab environment based on knowledge discovery techniques using open source tools. This research used I 195 datasets in the experiment wherein dynamic analysis and security metrics have been applied. From the experimentalresult, the FGA technique has produced an overall detection accuracy rate of 99.749 °o with 0.014 % false positive rate, thus outperforming the existing work of the OlexGA technique with 4I. 05 % improvement. This research has produced a technique that can detect and respond to cloud worm attacks. |
|---|