模块化免疫神经网络模型在计算机病毒分类检测中的应用

结合神经网络优化、聚类的特点、免疫网络识别自己、非己和抗体处理抗原的机理以及遗传算法中适者生存和优胜劣汰的原理，探讨一种基于神经网络方法下的新的计算机病毒分类检测模型，该模型能动态实现系统处理信号的分类，而且对每类信号也给出了综合有效的处理方法，其利用神经网络和免疫网络相互作用的规律，最终形成一个动态的、自适应的、分布式的模型。该模型可有效地解决病毒分类检测问题，有较大潜力和发展空间。

Application of Modular Immune Neural Network in Classification and Detection of Computer Virus

The paper introduces a novel classification and detection model of computer virus based on neural network, with neural network＇s characteristics of clustering optimization, the principle of survival of the fittest in genetic algorithm, the ability of recognitions of self or nonself and mechanism of antibody treating antigen. The model realized the classification of system＇s signal, and the model effectively solved the problem of the classification and detection of computer virus. Finally, using the law of interaction between neural network and immune network, a dynamic, self-adaptive and distributed model is built, which has greater potential and more development space.