作 者: (刘达);
机构地区: 国网湖南省电力公司技术技能培训中心,湖南长沙410131
出 处: 《网络空间安全》 2017年第8期32-34,共3页
摘 要: 在网络信息安全体系中,数据库的安全保护最为关键,入侵检测系统(IDS)是目前较为理想的数据库安全工具。论文提出了一种数据库入侵检测方法,包含一种可反映同一个数据库事务中SQL语句的关联性的数据库日志数据结构,以及采用朴素贝叶斯分类算法(NBC)的机器学习算法与入侵检测算法。实验证明,论文提出的检测系统具备较高的准确率。 The security of Database Management System is crucial for Information Security. An intrusion detection system(IDS), is the ideal solution to defend databases from intruders. This paper suggests an This paper suggests an anomaly detection approach that using the raw transactional SQL queries into a compact data structure and modelutilizes naive Bayes classifier(NBC) as the learning technique for creating profiles and detecting intrusions. Experimental results show the performance of the proposed model is outstanding.