将IFS理论引入信息安全领域,提出一种基于直觉模糊推理的入侵检测方法.首先,描述了入侵行为的特征属性、入侵检测的不确定性,以及现有入侵检测方法的特点与局限性,给出了原始数据预处理的方法.其次,将入侵特征属性直觉模糊化,建立了特征属性的直觉模糊集合及其隶属函数与非隶属函数.再次,建立了系统推理规则,设计了推理算法和清晰化算法.最后,选择KDDCUP99的入侵检测数据集,验证了方法的有效性.
The paper brings the theory of IFS into the field of information security and proposes a technique for intrusion detection based on intuitionistic fuzzy reasoning. First, intrusion features, nondeterminacy to intrusion detection, and the properties and vulnerabilities of the existing intrusion detection methods are analyzed and a method for data preprocessing in intrusion detection is given. Then, the intrusion features are transformed to intuitionistic fuzzy sets and the membership and nonmembership functions are devised. Subsequently, the inference rules of the system are constructed, the algorithms for reasoning and defuzzification are also devised. Finally, the validity is checked by providing intrusion detection instances with KDD CUP 99 dataset.