logpai/logparser
Logparser is a machine learning toolkit that provides automated log parsing and benchmarks for structured log analytics via event template extraction.
Awesome AI for Infra › Log Analysis & Intelligence
logpai/Log3C
Log3C is a research-backed, general framework designed to identify impactful service system problems from IT system logs. It leverages machine learning techniques to process both system log data and Key Performance Indicator (KPI) metrics, aiming for prompt and precise problem identification. The framework consists of four main steps: Log Parsing, Sequence Vectorization, Cascading Clustering, and Correlation Analysis. The core innovation lies in its cascading clustering algorithm, which efficiently groups large numbers of log sequence vectors by iteratively sampling, clustering, and matching. This approach allows Log3C to detect underlying system anomalies and correlate them with operational issues. Developed as a joint work between CUHK and Microsoft Research, Log3C is suitable for researchers and practitioners looking to implement advanced log analysis for AIOps.
https://github.com/logpai/Log3C
Also in Log Analysis & Intelligence
logpai/loglizer
Loglizer is an open-source machine learning toolkit designed for automated anomaly detection in system logs, supporting various supervised and unsupervised models.
logpai/Drain3
Drain3 is an online log template miner that extracts structured templates from raw log messages for enhanced observability and anomaly detection.