摘要
Abstract
Log data,as one of the most important data resources in software systems,records de-tailed information during system operation,and automated log anomaly detection is crucial for maintain-ing system security.With the widespread application of large language models in the field of natural language processing,Transformer-based log anomaly detection methods have been widely proposed.Traditional Transformer-based methods struggle to capture the local features of log sequences.To ad-dress this issue,this paper proposes a log anomaly detection method,LogTC,based on Transformer and Text-CNN.Firstly,logs are converted into structured log data through rule matching,while pre-serving the effective information in log statements.Secondly,log statements are divided into log se-quences using fixed windows or session windows according to log characteristics.Thirdly,natural lan-guage processing technology,specifically Sentence-BERT,is used to generate semantic representations of log statements.Finally,the semantic vectors of the log sequences are input into the LogTC log anom-aly detection model for detection.Experimental results show that LogTC can effectively detect anoma-lies in log data and achieves good results on two datasets.关键词
日志异常检测/深度学习/词嵌入/Transformer/Text-CNNKey words
log anomaly detection/deep learning/word embedding/Transformer/Text-CNN分类
计算机与自动化
