In order to accurately determine the vulnerability situation of electric power IoT terminals,an automatic vulnerability mining method for electric power IoT terminals based on improved ASTNN network is proposed.Historical vulnerability information of electric power IoT terminals is collected,and the features of these vulnerability information are extracted as the basis for automatic mining.Combine multiple terminal vulnerability detection methods to establish an anti-filtering rule set that can be used for vulnerability analysis.The neural network based on abstract syntax tree is analyzed,and the expression subtree is trimmed at the coding layer to form an improved ASTNN network that can better understand the semantic information,based on which the automatic mining framework is built to realize the vulnerability mining of power IOT terminals.The experimental results show that the F1 value of the proposed method is always stable above 0.97,which meets the requirements of vulnerability detection for power IoT terminals.The method improves the accuracy and efficiency of vulnerability mining and has greater application value.
关键词
改进ASTNN网络/电力物联网/终端漏洞/深度学习/反过滤规则/自动挖掘
Key words
improved ASTNN network/PowerInternet of Things/Terminalvulnerability/Deep learning/Anti filtering rules/Automatic mining
维普
万方数据