信息安全研究2026，Vol.12Issue(4)：383-392,10.DOI:10.12379/j.issn.2096-1057.2026.04.11
融合语义特征的日志异常检测方法研究
Research on Log Anomaly Detection Method Integrating Semantic Features
摘要
Abstract
With the continuous expansion of system functionalities,the volume of system logs has grown exponentially,presenting substantial challenges to conventional anomaly detection approaches.Deep learning-based log anomaly detection techniques have gradually become a research hotspot due to their powerful feature extraction capabilities.This study proposes a semi-supervised log anomaly detection model LogSem,which integrates semantic features.By introducing log content vectors that contain semantic information of the main log content and incorporating masked log key prediction tasks and hypersphere volume minimization tasks for semi-supervised learning,the model deeply explores the semantic features of logs.Experiments conducted on three mainstream datasets show that the proposed method outperforms the LogBERT baseline model in terms of the F1 score.Furthermore,this study explores and verifies the feasibility of addressing the out-of-vocabulary problem through semi-supervised learning.关键词
日志异常检测/日志解析/深度学习/半监督学习/BERT模型Key words
log anomaly detection/log analysis/deep learning/semi-supervised learning/BERT model分类
信息技术与安全科学引用本文复制引用
陈瀚文,章乐,池亚平,姜波,王志强..融合语义特征的日志异常检测方法研究[J].信息安全研究,2026,12(4):383-392,10.基金项目
中央高校基本科研业务费专项资金项目(3282024050) （3282024050）
国家重点研发计划项目(2023YFC2206402) （2023YFC2206402）
