Threat detection based on multi-scale spatiotemporal feature fusion of user behavior
Xianggen Wang, Fuxi Wang, Jiajia Cui, Bo Shen · 2023
With the great changes brought about by digitization, traditional security threat detection capabilities have been greatly challenged. Traditional threat detection technologies are based on signatures, rules and manual analysis, and there are serious lags and blind spots in security visibility. Unknown attacks cannot be detected and are easily bypassed. Multi-scale user behavior fusion analysis uses artificial intelligence methods and spatiotemporal feature engineering to associate multisource heterogeneous user behavior feature data to realize threat detection of multi-modal and multi-scale data.