基于流记录偏好度的多分类器融合流量识别模型traffic classification .pdfVIP

第34 卷第10 期 通 信 学 报 Vol.34 No. 10 20 13 年10 月 Journal on Communications October 20 13 基于流记录偏好度的多分类器融合流量识别模型 1,2,3 1,2 董仕 ，丁伟 (1.东南大学 计算机科学与工程学院,江苏 南京 211189; 2. 东南大学计算机网络和信息集成教育部重点实验室,江苏 南京 211189; 3.周口师范学院计算机科学与技术学院,河南 周口 466001) 摘 要：通过将证据理论引入到流量分类的决策模块中，提出了偏好性和时效性权值，并通过实测数据对多分类 器识别模型进行验证，其结果表明该模型较好的克服了单分类器的片面性，通过对多个证据的融合来优化识别的 结果。 关键词：多分类器融合；证据理论；偏好性；机器学习 中图分类号：TP393.08 文献标识码：A 文章编号：1000-436X(2013) Traffic classification model based on fusion of multiple classifiers with flow preference DONG Shi1, 2, 3, DING Wei1, 2 (1.School of Computer Science and Engineering，Southeast University, Nanjing 211189 China; 2. Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, Nanjing 211189, China; 3. School of Computer Science and Technology ，Zhoukou Normal University ，Zhoukou 466001, China) Abstract: Introduced the concept of multi-classifier fusion which can improve the classification accuracy and overcome the disadvantage of single classifier. This paper introduced DS theory into decision module of traffic classification and proposed preference and timeliness. After analyzing multi-classifier model by simulation, the results show the new classifier model can overcome one sidedness of single classifier, depend on multiple evidences to optimize the traffic results. Key words: Multi-classifier; DS theory; Preference; Machine lear