Incremental learning traffic anomaly detection method based on deep learning

Abstract

The embodiment of the invention provides an incremental learning traffic anomaly detection method based on deep learning, and the method comprises the steps: collecting network traffic data, and carrying out the preprocessing of the network traffic data, and obtaining the processed network traffic data; performing feature pre-extraction on the processed network traffic data by using an LSTM model; establishing a decision tree improved support vector machine model, selecting optimal parameters through a folding cross validation algorithm, and constructing an optimal model; and training to obtain an improved vector machine model which is used for classifying the extracted network traffic data features and evaluating the classification effect. According to the method, a decision tree and SVM combined method is provided, LSTM is adopted to extract traffic information features according to network traffic data features, and then classification detection of abnormal traffic is carried out by using a model. Classification detection is successfully carried out based on the improved model, and compared with a traditional traffic anomaly detection method, a better classification effect is achieved.

Description

technical field [0001] This application relates to the direction of traffic anomaly detection, in particular to the incremental learning traffic anomaly detection method based on deep learning. Background technique [0002] With the rapid development and popularization and application of modern information technologies such as cloud computing, mobile Internet, and Internet of Things, big data has gradually participated in all aspects of economy, society, and technology. Big data has the characteristics of massive volume, high growth rate and diversification, and cannot be mined, analyzed and processed with general software tools within a limited time frame. When dealing with big data, a new model is adopted, which enables big data to support super-powerful decision-making, insight and processing capabilities, and bring various application convenience services to society and life, but new risks and challenges also arise immediately. Come. How to effectively avoid risks and ...

AI Technical Summary

Problems solved by technology

Hassan et al. proposed a hybrid deep learning model of convolutional neural network and weight reduction. According to the author's experiment, the model has a shorter time-consuming and higher accuracy rate from the results, but the false alarm rate is also lower. high

Images