| 基于PrefixSpan的应用层协议特征串提取算法 |
| |
| 引用本文： | 秦兆文,刘嘉勇.基于PrefixSpan的应用层协议特征串提取算法[J].信息安全与通信保密,2014(6):105-108. |
| |
| 作者姓名： | 秦兆文 刘嘉勇 |
| |
| 作者单位： | 四川大学电子信息学院信息安全研究所,四川成都610064 |
| |
| 摘 要： | 根据经典数据挖掘算法PrefixSpan对应用层数据流净载荷进行投影挖掘,提取出可以唯一识别该应用的特征字符串。针对特征字符串有偏移的情况采用连续前缀投影,在生成频繁序列的过程中,针对字符串的基本特征加以约束条件,减少投影数据库的产生,提高特征字符串提取的效率。
|
| 关 键 词： | PrefixSpan算法 应用识别 特征字符串 前缀投影 |
A PrefixSpan-based Extraction Algorithm for Application Layer Protocol |
| |
| QIN Zhao-wen,LIU Jia-yong.A PrefixSpan-based Extraction Algorithm for Application Layer Protocol[J].China Information Security,2014(6):105-108. |
| |
| Authors: | QIN Zhao-wen LIU Jia-yong |
| |
| Institution: | (Institute of Information Security Electronic Information College,Sichuan Univ. , Chengdu Sichuan 610064, China) |
| |
| Abstract: | Based on the classical data mining algorithms PrefixSpan, the application layer data is projected to find the application identification character string. A serial prefix projection is used to solve the character string offset in the data flow. In the generating process of frequent sequence, some constraint condition is added to reduce the generation of projection database and improve the efficiency of character-string extraction. |
| |
| Keywords: | PrefixSpan application identification character string prefix projection |
| 本文献已被 维普 等数据库收录！ |
