作者: Zhenping Shi , Jie Li , Chentao Wu , Jinyuan Li
DOI: 10.1109/HPCC/SMARTCITY/DSS.2019.00335
关键词: Traffic volume 、 Artificial intelligence 、 Intrusion detection system 、 Anomaly detection 、 Computer science 、 Artificial neural network 、 Deep learning 、 Data mining
摘要: With the explosion of network traffic volume, high efficient and large-scale anomaly detection methods becomes necessary. However, existing often fail to take into account both delay accuracy. We propose a novel method, focusing on period-wise detection. use Long Short-Term Memory (LSTM) establish abnormal model. Besides, some big data processing frameworks for online collection preprocessing. Performance evaluation shows that our model outperforms other based traditional methodologies.