计算机免疫智能的连续应答机制及其应用研究

The development of artificial intelligence is in full swing, and its research is based on data and intelligent algorithms, meanwhile, it ignores the biological intelligence is a collaborative mechanism. Sequential Immune Responses (SIR) mechanism is a new body immune theory, which enables innate immunity to cooperative with antigens by the immune cells from the perspective of immune response. This project draws on the results of this bio-immune research, focusing on the anomaly detection of computer systems, the balance theory based on SIR and the adaptive and adjustment model. The main contents of this project include: (1) The formal description of feature space and the construction of feature tree; (2) Find out the anomaly by using the method of differential description of variation tendency in mathematics; (3) Combine the immune nonlinear model and SIR mechanism to construct the SIR hierarchical collaboration model in computer fields, including basic abnormal detection model, natural killer cell model, macrophage model and danger theory model;(4) Through the hierarchical collaboration model, the concentration of antigen in the feature space library is adjusted by feedback, so as to achieve the adaptability and collaboration of the model. The above researches may be instructive to the practical study of computer immune in computer system security, network security situational awareness, earthquake prediction, intelligence analysis and complex information processing.

人工智能发展如火如荼，而其研究是基于数据和智能算法的，忽略了生物智能中的协同和机制特性。连续免疫应答机制是一种新的机体免疫理论，该机制使得先天免疫能够从应答的角度出发，协调免疫细胞进行应答。本项目借鉴此生物免疫研究成果，重点研究计算机系统的异常发现方法、基于连续免疫应答机制的平衡理论和自适应调节模型。项目主要研究内容包括：（1）特征空间的形式化描述和特征树的构建；（2）借鉴数学中微分描述变化趋势的方法发现异常；（3）结合免疫的非线性模型和连续免疫应答机制构建计算机连续免疫应答层次协作模型，包括基础异常发现模型、自然杀伤细胞模型、巨噬细胞模型和危险理论模型；（4）通过层次模型协作对特征空间库中抗原的浓度进行反馈调节，以达到模型的自适应性和协同性。上述研究对计算机免疫在计算机系统安全、网络安全态势评估、地震预测、情报分析和复杂信息处理的实用化有较强的推荐作用。

人工智能发展如火如荼，而其研究是基于数据和智能算法的，忽略了生物智能中的协同和机制特性。连续免疫应答机制是一种新的机体免疫理论，该机制使得先天免疫能够从应答的角度出发，协调免疫细胞进行应答。本项目借鉴此生物免疫研究成果，重点研究计算机系统的异常发现方法、基于连续免疫应答机制的平衡理论和自适应调节模型。项目主要研究内容包括：（1）特征空间的形式化描述和特征树的构建；（2）借鉴数学中微分描述变化趋势的方法发现异常；（3）结合免疫的非线性模型和连续免疫应答机制构建计算机连续免疫应答层次协作模型，包括基础异常发现模型、自然杀伤细胞模型、巨噬细胞模型和危险理论模型；（4）通过层次模型协作对特征空间库中抗原的浓度进行反馈调节，以达到模型的自适应性和协同性。上述研究对计算机免疫在计算机系统安全、网络安全态势评估、地震预测、情报分析和复杂信息处理的实用化有较强的推荐作用。

内容获取失败，请点击重试