Computer network security risk evaluation

Inspired by the relationship between antibody concentration and intrusion intensity of pathogens in the human immune system, here we demonstrate that the way, which a doctor diagnoses a patient by checking the patient's temperature, can be exploited to evaluate the security risk faced by a computer network. In this model, which we call network temperature evaluation, the system itself learns what attacks it suffers, where the intrusions occur, whether the intrusions are serious or not. In addition, the system learns what the current popular attacks are, and where the most serious disastrous area is. With the proposed techniques, all the above information can be calculated quantitatively and in real-time.