Research On Network Traffic Classification Technology Based On Support Vector Machine

With the development of network technology,the traffic in the network has grown extremely rapidly.Accurate and timely traffic classification is critical in network security and traffic engineering.For the growing use of network applications,traditional methods based on port numbers and protocols are particularly inefficient.This paper further studies how to use the machine learning method to improve the classification accuracy of network traffic and how to improve the real-time transmission capability when dealing with large-scale traffic data.First of all,when extracting traffic characteristics,it faces a variety of traffic characteristics,which will increase the overhead of training time.In order to solve the problem of large support cost of support vector machine due to feature redundancy,this paper proposes a feature extraction method based on maximum correlation minimum redundancy.The method uses the mutual information between random variables to extract the eigenvalues with the least correlation minimum redundancy.Simulation results show that this method can effectively reduce training time and improve training efficiency.Secondly,in order to reduce the impact of noise on the classification of SVM,this paper proposed a new fuzzy factor-based SVM traffic classification algorithm.This algorithm mainly calculates the membership degree of each sample corresponding to its type,and the membership degree is determined by the distance from the sample to the hyperplane,and is used to measure the weight of each sample.The simulation results show that this method can effectively reduce the impact of noise and outliers on classification accuracy,thus improving the accuracy of SVM classification.Finally,based on feature selection and SVM algorithm research,a Spark-based directed acyclic graph SVM traffic classification algorithm proposed in this paper.The Spark based memory distributed computing is used to implement the directed acyclic graph SVM algorithm,and finally the parallel network traffic classification model is obtained.Classification efficiency of network traffic.The results show that the network traffic classification model based on SVM designed in this paper can effectively improve the classification efficiency of network traffic and improve the accuracy of traffic classification.