Detection method and system based on SRU network abnormal traffic

The invention relates to a detection method and system based on SRU network abnormal traffic. Attribute mapping is carried out on traffic data, so that the to-be-detected data meets a model data format requirement; and the traffic data is subjected to code conversion and dimension reduction processing, the SRU network is used for feature learning, and a classifier is trained to realize abnormal traffic classification. The problem that an existing abnormal traffic detection algorithm based on deep learning is low in training speed is solved. 本发明涉及一种基于SRU网络异常流量的检测方法及系统。通过对流量数据进行属性映射，使待检测数据符合模型数据格式需求；通过对流量数据进行编码转换、降维处理并通过使用SRU网络进行特征学习，并训练分类器实现异常流量分类。本发明解决现有的现有基于深度学习的异常流量检测算法训练速度慢的问题。