2022 年第 2 期 等级保护 doi :.1671- 基于 LSTM-Attention 的内部威胁检测模型 aracteristics and psychological data were extracted to describe the daily activities of users. Secondly, the long short-term memory (LSTM) network and the attention mechanism were used to learn the user’s behavior pattern, and calculate the deviation between the real behavior and the predicted behavior. Finally, multilayer perceptron was used to make comprehensive decisions based on these deviations to identify abnormal behaviors. Experimental results on the CERT insider threat dataset show 收稿日期:2021-08-04 基金项目:国家自然科学基金 [62072239];国家重点研发计划 [2018YFB0804701];河北省科技厅科技计划 [20377725D] 作者简介:张光华(1979—),男,河北,教授,博士,主要研究方向为网络与信息安全;闫风如(1997—),女,河北,硕士研究生, 主要研究方向为网络与信息安全;张冬雯(1964—),女,河北,教授,博士,主要研究方向为网络与信息安全;刘雪峰(1985—),男, 安徽,副教授,博士,主要研究方向为隐私保护。 通信作者:张冬雯 ******@ 1等级保护2022 年第 2 期 that the proposed ITDBLA model achieves an AUC score of , which show a stronger ability to learn user activity patterns and detect abnormal behaviors. Key words: LSTM; attention mechanism; user and entity behavior analysis; insider threat d