EagerLog: Active Learning Enhanced Retrieval Augmented Generation for Log-based Anomaly Detection
Authors:
- 6 April 2025
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abstract
Logs record essential information about system operations and serve as a critical source for anomaly detection, which has generated growing research interest. Utilizing large language models (LLMs) within a retrieval-augmented generation (RAG) framework for log-based anomaly detection is an effective approach due to its strong generalization capabilities and efficient few-shot performance. However, the effectiveness of this method hinges on the quality of the knowledge source, which can be impacted by noise and changes within the software systems. Facing these problems, in this paper, we propose a novel log-based anomaly detection method named EagerLog, employing active learning to choose the logs for humans to label, thereby adding them to the knowledge source, thus enhancing the knowledge source and maintaining its quality. Our experiments on three open datasets (BGL, Thunderbird, Zookeeper) and one industrial dataset demonstrate that EagerLog can achieve 93.65% F1 score with approximately 10 labeled log sequences, surpassing existing methods by 15.32%.Keywords
This publication has 25 references indexed in Scilit:
- LogGPT: Exploring ChatGPT for Log-Based Anomaly Detection2023
- AcLog: An Approach to Detecting Anomalies from System Logs with Active Learning2023
- LogFlash: Real-time Streaming Anomaly Detection and Diagnosis from System Logs for Large-scale Software Systems2021
- Semi-Supervised Log-Based Anomaly Detection via Probabilistic Label Estimation2021
- LogTransfer: Cross-System Log Anomaly Detection for Software Systems with Transfer Learning2020
- LogDC: Problem Diagnosis for Declartively-Deployed Cloud Applications with Log2017
- LogSed: Anomaly Diagnosis through Mining Time-Weighted Control Flow Graph in Logs2017
- Drain: An Online Log Parsing Approach with Fixed Depth Tree2017
- An Approach for Anomaly Diagnosis Based on Hybrid Graph Model with Logs for Distributed Services2017
- LOGAN: Problem Diagnosis in the Cloud Using Log-Based Reference Models2016