A Convolution Neural Network and Genetic Algorithm-Based Abnormity Detection Model for Cyber Attacks
Abstract
With the rapid development of the Internet and network technology, network security has become increasingly prominent. Regarding the important issue of network attack detection, traditional methods often find it difficult to effectively capture and identify new types of network attack behaviors, while models based on deep learning and evolutionary algorithms are considered to better adapt to complex and ever-changing network attack environments. This paper aims to explore how to combine convolutional autoencoder (CAE) and genetic algorithm (GA) to construct an efficient network attack detection model and improve the ability of network security defense. First, the convolutional autoencoder is used to effectively learn the feature representation of network data, and combined with hierarchical attention mechanism, appropriate weights are assigned to classification tasks under different features and feature fusion is performed. Using GA to adaptively optimize random forest and improve its performance and robustness in network attack detection. Ultimately, achieving better network security protection results and effectively preventing and combating various network security threats. And conduct experimental verification on multiple datasets and compare with other benchmark methods. The results show that the model has achieved significant improvement in the detection of text network attacks, and can more effectively classify various types of attacks, bringing new technological breakthroughs and application prospects for network attack detection. It will also provide new ideas and methods for the further development of network security, effectively ensuring the safe and stable operation of network systems.
This paper was recommended by Regional Editor Takuro Sato.
