Method and device for constructing feature recognition library and flow recognition method

Abstract

The invention relates to a feature recognition library construction method and device, a traffic recognition method, computer equipment and a storage medium. The feature recognition library construction method comprises the steps of obtaining first traffic and a message set obtained by analyzing the first traffic; according to the classification field of each message in the message set, dividing the traffic with the same classification field in the first traffic into the same set; according to the transmission parameter information of each message in the message set, clustering the traffic with the same classification field in the same set to obtain a plurality of traffic clusters; obtaining a second flow and a function category corresponding to the second flow; according to the similarity between the second traffic and each traffic cluster in the plurality of traffic clusters, extracting identification features corresponding to the function categories from the traffic clusters; and constructing a feature recognition library according to the recognition features and the function categories corresponding to the recognition features. By adopting the method, the accuracy of feature recognition in the feature recognition library can be improved, so that the accuracy is higher when the feature recognition library is adopted for traffic recognition.

Description

technical field [0001] The present application relates to the technical field of traffic identification, in particular to a method and device for constructing a feature identification library, a traffic identification method, computer equipment and storage media. Background technique [0002] With the rapid development of Internet technology, the number of network applications is also growing rapidly. The development of network applications brings great convenience to people's life, but the complexity and diversity of network applications also brings great challenges to network application management and traffic control. In order to effectively cope with the challenges brought by the rapid development of network applications, real-time and accurate traffic identification research has become one of the important research issues in the current network management research field. [0003] The traditional traffic identification method is to identify based on the payload, and det...

AI Technical Summary

Problems solved by technology

[0004] However, the traditional technology lacks the distinction of network traffic, resulting in low accuracy of identifying features in the feature recognition database, so the effect of traffic recognition is not ideal

Images