Research On Network Information Security Recognition Based On Neighborhood Rough Decision Theory

Under the development trend of "Internet +",traditional industries are moving in the direction of informatization and digitization,and network information is gradually flooding people’s lives.The popularity of the Internet has changed people’s way of life to a large extent,enriched our daily life,and brought huge social and economic benefits.At the same time,more and more network information security problems have gradually surfaced,the new normal of the Internet is facing more severe challenges,and the importance of network information security identification and early warning mechanisms has become increasingly prominent.This paper takes advantage of the neighborhood rough set’s ability to effectively deal with mixed data of both symbolic and numerical types,seeks the foundation of network information security from multiple perspectives,and provides decision-making basis for my country’s network information security identification behavior and network information security risk early warning.Aiming at the network information security data with wide sources and wide coverage,this paper proposes a neighborhood multi-granularity decision rough set for hybrid information system after comprehensively evaluating various rough set models and considering the relative distance between the data.After that,a complete definition of related metrics is given,important properties of the model are discussed,attribute reduction methods and decision rules are generated,and the validity of the model is verified by examples.Starting from the current situation of network information security,by sorting out all kinds of special network information security incidents,summarizing the causes of network information security incidents and indicating the direction of identification,according to relevant principles,constructing from four levels of technology,management,network equipment,and external environment.Network information security identification factor index system,and detailed indicators.Taking appropriate network information samples as an example,a decision information system is constructed.After the corresponding results are obtained by using the improved rough set model,attribute reduction and rule extraction are carried out,and the final decision rules generated can effectively improve the identification efficiency of network information security.It can be seen from examples that the improved neighborhood rough set model has stronger applicability,can well control the network information security identification system,and provide ideas for reasonably controlling network information security risks.