KMApriori：一种有效的数据库异常检测方法

本文提出了一个基于聚类分析与关联规则挖掘技术的有效数据库异常检测KMApriori方法，设计了一个系统框架，研究了该框架中的关键技术。通过实验表明，KMApriori方法能够检测出数据库用户的异常行为。与基于关联规则挖掘技术的数据库异常检测Apriori方法相比，在时间开销上，KNLkpriori方法优于Apriori方法，并且随着数据量的增加加，KMApriori的优越性更明显;此外，KMApriori方法检测正确率也高于Apriori方法。

KMApriori:An Efficient Anomalous Detection Approach to Database Systems

Based on clustering analysis and associate rule mining, an efficient anomalous detection approach is presented, and a database anomalous detection architecture is designed. The critical technology of the architecture is also studied in this paper. KMApriori, the database anomaly detection algorithm which is presented in this paper, can detect the anomalous operations of users, and is superior to the database anomaly detection algorithm Apriori which is based on associate rule mining. The more database records, the more distinctness of the superiority. Furthermore, the detection rate of KMApriori is also higher than Apriori.