The Study Based On Danger Model Inspired Algorithms

Lately, immune system has been a novel subject of biologic information. It is another special system which can identify self and non-self(SNS) and response exactly .Then save the functional system. After the emergence of SNS model there is another new model go around in immunology. That is the danger model. Matzinger who is the main introducer of this new theory believed that immune system can not to identify self and non-self but to distinguish whether have the danger signal. So danger model question the traditional immune model (SNS). Using danger theory many aspects of medicinal phenomena can explain perfectly. Nowadays,this novel theory arouse many experts'interesting.The theory of danger model is not the extended methods of immune response but a novel theory model. Since now though many immunologists did not authorize this new theory. But at the field of artificial immune system we can extend this kind of model in not take the shackle of immune system into account.Based on these developments, many good algorithms for various data mining tasks are obtained.Based on previous research, the major research works in this article are:1 search of the theory of danger model inspired algorithms has been proposed. And try to create a novel a model of danger model inspired algorithms then prove the feasibility of the algorithms through out experiments.2 Then introduce this new algorithms to the areas of data mining .The major applications in data mining involve Association Rule Mining and data cluster analysis.3 In the later time the trial research of safety aspects based on danger model inspired algorithms has been proposed. Then in the paper we bring forward a adaptive mailbox and intrusion detection system based on danger model inspired algorithms.4 Finally, this paper try to set up a web-safety system using danger model inspired algorithms.