基于RIPPER的网络流量分类方法
发布时间：2018-09-17 20:33
【摘要】：利用一种规则学习方法中的重复增量式降低错误剪枝方法解决网络流量分类问题。利用该方法能够挖掘出网络流属性特征和类别之间的相关关系,并将挖掘出的关系构成分类器用于网络流量分类。该方法能够解决传统机器学习方法在网络流量中有大量的不平衡数据集时,分类错误率高等问题。实验证明,该方法在网络流量分类标准数据集上具有很高的分类准确率、查全率和查准率。
[Abstract]:The problem of network traffic classification is solved by reducing error pruning in a rule learning method. By using this method, the correlation between the attribute characteristics of network flow and the class can be mined, and the relationship constructed by this method can be applied to the classification of network traffic. This method can solve the problem of high classification error rate when the traditional machine learning method has a large number of unbalanced data sets in network traffic. Experiments show that this method has high classification accuracy recall and precision on the standard data set of network traffic classification.
【作者单位】： 哈尔滨理工大学计算机科学与技术学院;
【基金】：国家自然科学基金(60903083,61502123) 黑龙江省新世纪人才项目(1155-ncet-008) 黑龙江省博士后科研启动基金
【分类号】：TP393.0
本文编号：2247011
[Abstract]:The problem of network traffic classification is solved by reducing error pruning in a rule learning method. By using this method, the correlation between the attribute characteristics of network flow and the class can be mined, and the relationship constructed by this method can be applied to the classification of network traffic. This method can solve the problem of high classification error rate when the traditional machine learning method has a large number of unbalanced data sets in network traffic. Experiments show that this method has high classification accuracy recall and precision on the standard data set of network traffic classification.
【作者单位】： 哈尔滨理工大学计算机科学与技术学院;
【基金】：国家自然科学基金(60903083,61502123) 黑龙江省新世纪人才项目(1155-ncet-008) 黑龙江省博士后科研启动基金
【分类号】：TP393.0
【相似文献】
相关期刊论文 前1条
1 冀汶莉;;基于RIPPER短序列匹配算法在入侵检测中的优化研究[J];兰州工业高等专科学校学报;2007年01期
相关硕士学位论文 前2条
1 张灿;基于规则的论坛爬取与抽取一体化[D];华东师范大学;2011年
2 吉伟锋;基于汽车行业的Web信息抽取研究[D];辽宁工程技术大学;2007年
，本文编号：2247011
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2247011.html
