Network Anomaly Detection Algorithm Based on Deep Learning and Data Mining

Author:
Open Access
Publisher website
Google Scholar
Abstract
In this paper, we propose a new detection method for network anomaly detection, a key problem in the field of cyber security, which combines deep learning feature extraction and DBSCAN clustering algorithm. First, in terms of data processing and feature learning, we employ deep learning models CNN and RNN to automatically extract useful features from complex network traffic data. These advanced features can represent network behavior more accurately and provide richer information for subsequent clustering and anomaly detection. Then, using the DBSCAN algorithm, we can not only effectively identify normal behavioral patterns, but also accurately detect anomalies and potential threats. In the experimental part, by evaluating on the publicly available KDD Cup 99 dataset, the method in this paper demonstrates its superiority in key metrics such as accuracy and recall.

This publication has 5 references indexed in Scilit: