Internet business data classification method and system

Abstract

The invention provides an internet service data classification method. The method comprises the following steps: determining and capturing a to-be-classified data packet; inputting the to-be-classified data packet into a pre-trained data classification model to obtain a data classification result, wherein the data classification model is obtained by adopting an SVM algorithm for training based on Internet data packet preset features. According to the data classification method based on the SVM classification algorithm and taking the statistical characteristics of the data packets as the standard, the SVM algorithm is proved to perform multi-category classification processing on the captured data packets by taking the packet length of the data packets and the time interval between the packets as the classification characteristics under the Ethernet network condition; and the classification effect can be further optimized by adjusting related parameters of the SVM algorithm.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to a method and system for classifying Internet service data. Background technique [0002] With the development of the Internet, the composition of network traffic is becoming more and more complex, and the total amount of network traffic is also increasing. In the context of big data, the identification and classification of Internet data is of great significance to improving the level of network communication, ensuring network security and optimizing network service quality. [0003] Existing researches have realized the use of various algorithms to classify various levels of network traffic. Since the data streams on the Internet have the characteristics of various types and large quantities, the algorithms for classifying Internet data in reality are important to ensure the classification accuracy. The premise is mostly more complex. For example, some scholars have studi...

AI Technical Summary

Problems solved by technology

Although these methods can achieve better classification results, they also have some disadvantages: the complexity of the algorithm is high, a large number of parameters are required to be adjusted, and more data features are required as classification standards; and the classification time is long, which cannot meet the requirements of Internet data classification. Requirements for classification efficiency in this scenario

Images