Designing a new model for worm response using security metrics

Nowadays, worms are becoming more sophisticated, intelligent and hard to be detected and responded than before and it becomes as one of the main issues in cyber security. It caused loss millions of money and productivities in many organizations and users all over the world. Currently, there are many works related with worm detection techniques but not much research is focusing on worm response. Therefore, in this research paper, a new model to respond to the worms attack efficiently is built. This worm response model is called as eZSiber, inspired by apoptosis or also known as cell-programmed death. It is a concept borrowed from human immunology system (HIS), where it has been mapped into network security environment. Once the users computer detects any indication of the worm attacks, the apoptosis is triggered. In order to trigger the apoptosis, security metrics plays a very important role in identifying the weight and the severity of the worm attacks. In this model, the static and dynamic analyses were conducted and the machine learning algorithms were applied to optimize the performance. Based on the experiment conducted, it produced an overall accuracy rate of 99.38 % using Sequential Minimal Optimization (SMO) algorithm. This performance criteria result indicated that this model is an efficient worm response model. Springer International Publishing Switzerland 2015.