N-Trans: Parallel Detection Algorithm for DGA Domain Names

Domain name generation algorithms are widely used in malware, such as botnet binaries, to generate large sequences of domain names which some registered by cybercriminals. Accurate detection malicious domains can effectively defend against cyber attacks. The the use traditional machine learning has been explored many researchers, but still is not perfect. To further improve on this, we propose a novel parallel model named N-Trans that based N-gram algorithm with Transformer model. First, add flag bits first and last positions for combination framework detect name. extract letter features capture position letters It relationship between letters. In addition, it accurately distinguish legitimate names. experiment, dataset legal Alexa collected 360 Security Lab. experimental results show achieves 96.97% accuracy DGA detection. identify outperforms mainstream algorithms.