流量异常检测中的直觉模糊推理方法

发布时间：2018-03-22 02:37

  本文选题：网络　切入点：信息安全　出处：《电子与信息学报》2015年09期 　论文类型：期刊论文

【摘要】：针对网络流量特征属性不确定性和模糊性的特点,将直觉模糊推理理论引入异常检测领域,该文提出一种基于包含度的直觉模糊推理异常检测方法。首先设计异常检测中特征属性的隶属度与非隶属度函数,其次,给出基于包含度的强相似度计算方法并生成推理规则库,再次给出多维多重式直觉模糊推理规则,最后建立异常检测中的直觉模糊推理方法。通过对异常检测标准数据集KDD99的实验,验证该方法的有效性,与常见经典异常检测方法对比,该方法具有更良好的检测效果。
[Abstract]:In view of the uncertainty and fuzziness of network traffic characteristics, the intuitionistic fuzzy reasoning theory is introduced into the field of anomaly detection. In this paper, an intuitionistic fuzzy reasoning anomaly detection method based on inclusion degree is proposed. Firstly, the membership and non-membership functions of feature attributes in anomaly detection are designed. The strong similarity calculation method based on inclusion degree is presented and the inference rule base is generated. The multi-dimensional and multi-fold intuitionistic fuzzy reasoning rules are given again. Finally, the intuitionistic fuzzy reasoning method in anomaly detection is established. The effectiveness of this method is verified by the experiment of KDD99, the standard data set of anomaly detection. Compared with the classical anomaly detection method, this method has better detection effect.
【作者单位】： 空军工程大学防空反导学院;
【分类号】：TP393.06
，

本文编号：1646723