基于均衡化样本类别的图书馆网络入侵自动检测研究

Research on library network intrusion automatic detection based on balanced sample categories

针对当前图书馆网络入侵检测准确率,误报率高的问题,以获得更理想的图书馆网络入侵检测结果,设计了基于均衡化样本类别的图书馆网络入侵自动检测方法。首先捕获图书馆网络数据包,提取数据包中的图书馆网络入侵检测特征,然后通过均衡化样本类别算法合成入侵检测样本集合,并通过统计模型和邻居节点的实际活动参数实现入侵检测,最后进行了图书馆网络入侵检测仿真实验,结果表明,提高了图书馆网络入侵检测正确率,降低了图书馆网络入侵检测误报率,图书馆网络入侵检测效率高,具有较高的实际应用价值。

Aiming at the problems of high accuracy and false positive rate of network intrusion detection in libraries,in order to obtain more ideal results of library network intrusion detection,an automatic library network intrusion detection method based on balanced sample categories is designed.First,capture library network data packets,extract library network intrusion detection features from the data packets,then synthesize intrusion detection sample sets through a balanced sample category algorithm,and implement intrusion detection through statistical models and actual activity parameters of neighbor nodes.Finally,carry out library network intrusion detection simulation experiments,and the results show that this paper improves the accuracy of library network intrusion detection,It reduces the false alarm rate of library network intrusion detection.The library network intrusion detection has high efficiency and high practical application value.