Research And Application Of Intrusion Detection Algorithm Based On Rough Set Theory

Along with the development of Internet of things,cloud computing,big data,artificial intelligence and 5 G communication technology,network information security has become a hot issue,and the network intrusion detection technology is one of the key technologies to ensure network security.Because of the large amount of data and high dimension of network security intrusion detection,the traditional data analysis technology is difficult to obtain satisfactory results,so the network intrusion detection technology has become the focus area among experts and scholars.This paper proposes two methods of attribute reduction based on rough set theory for network intrusion detection.The first is RSDB algorithm,which sorts character attributes from large to small,selects the character attributes with the largest number of values to block the data,and reduces the attributes according to the feature attributes of data.The second is RE-RFE algorithm,which sorts the feature attributes obtained by the Relief-F and RFE algorithms,obtains their respective kernel attributes,and then obtains the intersection of the results of the two different algorithms,and calculates the final kernel attributes.The experimental results show that compared with the traditional data dimensionality reduction or attribute reduction algorithm,these two improved algorithms can help us to solve the problem that the data is difficult to analyze because of the large amount of data,the complex conditional attributes and the variety of attacks without distortion of the original data.The intelligent block algorithm based on rough set theory and efficient kernel algorithm have higher classification accuracy and faster average detection rate.