LogBERT：通过BERT进行日志异常检测
LogBERT: Log Anomaly Detection via BERT
摘要：
检测在线计算机系统中的异常事件对于保护系统免受恶意攻击或故障至关重要。系统日志记录了计算事件的详细信息，被广泛用于系统状态分析。在本文中，我们提出了LogBERT，日志异常检测的双向编码器表示从Transformers（BERT）的基础上的自监督框架。LogBERT通过两个新的自监督训练任务来学习正常日志序列的模式，并且能够检测出潜在模式偏离正常日志序列的异常。在三个日志数据集上的实验结果表明，LogBERT优于最先进的异常检测方法。
英文摘要：
Detecting anomalous events in online computer systems is crucial to protect the systems from malicious attacks or malfunctions. System logs, which record detailed information of computational events, are widely used for system status analysis. In this paper, we propose LogBERT, a self-supervised framework for log anomaly detection based on Bidirectional Encoder Representations from Transformers (BERT). LogBERT learns the patterns of normal log sequences by two novel self-supervised training tasks and is able to detect anomalies where the underlying patterns deviate from normal log sequences. The experimental results on three log datasets show that LogBERT outperforms state-of-the-art approaches for anomaly detection.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
