基于SVM的恶意流量检测及其改进方法分析  

Analysis of Malware Traffic Based on SVM and Its Improvement

作　　者：张远 徐逸卿[2] Zhang Yuan;Xu Yiqing(Nanjing Institute of Information Technology,Nanjing Jiangsu 210000,China;Nanjing Forestry University,Nanjing Jiangsu 210037,China)

机构地区：[1]南京信息技术研究院,江苏南京210000 [2]南京林业大学,江苏南京210037

出　　处：《信息与电脑》2018年第12期29-31,共3页Information & Computer

摘　　要：网络的迅猛发展给人们的生活带来了极大的方便,但同时也出现了恶意流量。如何识别恶意流量,近年来国内外许多学者对此问题进行了大量的研究和实践。现有的大多数处理方法都基于机器学习的流程识别技术。在这些机器学习方法中,支持向量机技术具有培训时间短、泛化能力强的优点。笔者分析了它的主要缺点,包括它需要大量的样本标签,导致高成本,并提出了一种基于支持向量机和协同训练的恶意流量检测方法。该方法通过引入Co-training半监督方法来减少样本标签的数量,保持分类的准确性。The rapid development of the Internet has brought great convenience to people＇s lives, but at the same time malicious traffic has also emerged. How to identify malicious traffic problems, many scholars at home and abroad in recent years have conducted a lot of research and practice on this issue. Most of the existing processing methods are based on machine learning flow recognition technology. Among these machine learning methods, the support vector machine technology has the advantages of short training time and strong generalization ability. The author analyzes its main shortcomings, including the need for a large number of sample labels to lead to high cost, and a malicious traffic detection method based on support vector machines and collaborative training is proposed. This method introduces Co-training senti supervised method to reduce the number of sample labels and keep the accuracy of classification.

关 键 词：恶意流量 支持向量机 协同训练 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]