中文摘要：

传统方法在对公共网络入侵数据检测时存在冗余度高、维数大、精确度差等问题。为了提高公共网络安全防护的实时性和有效性，提出一种基于优化粗糙集理论的公共网络检测方法。针对有入侵风险的数据进行检测和筛选，在粗糙集（RS）概念基础上对其精度进行优化，减少信息的丢失，运用MDLP运算准则完成对数据的离散化处理，使用遗传算法进行数据约简，导出数据分类规则并识别出入侵数据。仿真试验结果表明，所提出的入侵数据检测方法，在入侵检测率和误差率方面传统算法更为有效。

英文摘要：

Traditional method exists high redundancy,large dimension,poor accuracy and so on in the process of public network intrusion data detection. In order to improve the real?time performance and effectiveness of public network security pro?tection,a public network detection method based on the improved rough set theory is put forward to detect and screen the data which has invasion risk,optimize the detecting accuracy based on rough set concept,and reduce the information loss. TheMDLP operational criterion is adopted to complete the discretization processing of the data. The genetic algorithm is used to car?ry on the data reduction,derive data classification rules and identify the intrusion data. The simulation results show that the pro?posed intrusion data detection method is more effective in the aspects of intrusion detection rate and error rate in comparison with the traditional algorithm.