.png){kind=icon}
返回
Uit-DGAdetector: detect domains generated by algorithms using machine learning
.png){kind=icon}
PRE
AI
DOI：10.1007/s10586-024-04363-0.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Recent developments in information technology have brought numerous benefits but have also created risks for information security. One notable threat is the domain generated by the algorithm (DGA) technique used by botnets, which allows them to automatically generate and register multiple domains to evade detection and control from network security systems. To address this issue, we conducted research on a domain classification model specific to botnet-generated domains. We developed three domain classification models: bigrams, long short-term memory networks (LSTM), and a combination of LSTM and one-hot encoding. In this study, we implemented an ensemble model using a domain classification system, named UIT-DGADetector. To optimize the system, we employed Kafka to queue and streamline the requests, thereby reducing the load on the classification server. The deployed system operates well and achieves a high accuracy rate in predicting the domain types. However, this model still has limitations in predicting Word-based DGA botnets. The process must be optimized to reduce the waiting time in the queue. This study aims to contribute to network security and information protection, particularly by addressing the issue of DGA botnets.
Keyword:
Machine Learning
LSTM
Domain Generated by Algorithms
Botnet
DGA botnet
Kafka
期刊
C
IF:
4.1
论文数:
5.1K
被引数:
7.5K
机构
引用论文
分享收藏Adjuvant Therapy Choices in Patients With Resected Non-Small-Cell Lung Cancer: Correlation of Doctors' Treatment Plans and Relevant Phase III Trial Data术后非小细胞肺癌患者的辅助治疗选择：医生治疗计划与相关III期试验数据的关联
OAAI
OAAI分享收藏分享收藏分享收藏分享收藏A long short-term memory (LSTM)-based distributed denial of service (DDoS) detection and defense system design in public cloud network environment.png)
COMPUTERS & SECURITY
COMPUTERS & SECURITYIF5.4
PREAI
分享收藏
查看更多内容