【作者】 唐钒； 赵英； 郭倩玲； 滕建；

【Author】 TANG Fan;ZHAO Ying;GUO Qian-ling;TENG Jian;School of Information Science and Technology,Beijing University of Chemical Technology;Library,Beijing University of Chemical Technology;

【机构】 北京化工大学信息科学与技术学院； 北京化工大学图书馆；

【摘要】 网络技术的快速发展使网络流量的组成更加复杂,流量数据的分类问题面临着更大的挑战。文中提出一种基于卷积神经网络与长短时记忆循环神经网络的双分支网络分类方法,在双分支网络并行训练后确定相应的网络权值,拟合得出最终的分类结果。网络在普通流量、恶意流量以及加密流量的数据集测试中具有较高的准确率以及强健的稳定性。更多还原

【Abstract】 The rapid development of network technology makes the composition of network traffic more complicated and the classification of traffic data faces greater challenges.This paper proposes a two-branched network classification method based on the convolutional neural network and the long short-term memory recurrent neural network.After the parallel training of the two-branched network,the corresponding network weights are determined to fit the final classification result.The network has high accuracy and strong stability in the dataset test of ordinary traffic,malware traffic and encrypted traffic.更多还原