一种改进的模糊关联算法及其在IDS中的应用  

An Improved Algorithm of Fuzzy Association Rules and Its Application in IDS

作　　者：曾庆花[1] 王文国[1] 

机构地区：[1]曲阜师范大学计算机科学学院,山东日照276826

出　　处：《计算机技术与发展》2007年第7期236-239,共4页Computer Technology and Development

基　　金：国家人事部高层次留学人员回国工作资助项目(国人部发[2004]61号)

摘　　要：关联规则的发现是数据挖掘中的一个重要问题,但只是对离散型数据进行处理。为解决连续数量值属性的划分出现的“尖锐边界”问题,采用模糊划分,实现数据平滑过渡。由于入侵检测系统(IDS)对训练数据要求不高,文中提出了一种使用哈希链表改进模糊关联规则挖掘的新算法,且在挖掘过程中使用了等价类快速查找频繁项集,避免了反复扫描数据库及大量重复计算检验步骤。通过一个入侵检测系统的算例显示了其优越性,来提高对入侵数据的识别能力。Disoovery of association rule is an important problem in database mining, but it is merely used to handle the discrete data. To partition continuous quantitative attribute is handled by using fuzzy partition in order to solve the problem of sharpening boundary, which provides a smooth transition of data partition. On IDS the requirements of training data are very low. In the paper,an improved algorithm using Hashing tables on mining fuzzy association rules is proposed, and equivalence classes are introduced to search frequent itemsets quickly. With this algorithm the usual practice of repeatedly database scanning can be avoided. Its efficiency is showed with a typical use on intrusion detection system （IDS）from network datasets.

关 键 词：模糊关联 入侵检测系统 哈希链表 等价类 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]