Study Of Information Security Event Text Classification Method

In recent years,the rapid development of the Internet and cloud computing technology brings us a lot of information in the form of text,at the same time,the information contains a lot of useful knowledge,then the categorization of the information will become the primary problems and the basis of further research and analysis. On the other hand,information security incident occurred frequently, information security are increasingly important for individual and collective,sorting information security events can lay a foundation for future research.The paper analyzes the current standards of information security incidents and the classification system of information security event,collect the relevant sources of information,and sorting out a standard of information security event categories what is more suitable for scientific research.Then we in-depth research both the Chinese text classification steps,process and methods.We analyzed several common Chinese text classification algorithms and compared the effect of several kinds of algorithm for classification of information security events and proposed an improved method based on KNN algorithm.KNN algorithm is executed to compare all the training events every time,so that greatly reduce the execution efficiency of the algorithm,in order to effectively reduce the number of sample contrast test,we introduced the concept of lower approximation of rough set,the text under the approximate area will be considered to the class directly which significantly reduce the amount of calculation. According to the characteristics of the information security incidents,we can find their lower approximation area which can help us to improve the execution efficiency of the KNN algorithm and the accuracy and recall rate of algorithm.