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/loglizer
Loglizer provides a comprehensive machine learning-based toolkit specifically crafted for automated anomaly detection within system logs. It addresses the critical need in IT operations to monitor systems and identify abnormal behaviors and errors from the vast amounts of runtime information recorded in logs. The toolkit implements a variety of log analysis techniques, including both supervised models like Logistic Regression, Decision Trees, and SVM, as well as unsupervised models such as LOF, One-Class SVM, Isolation Forest, PCA, Invariants Mining, and Clustering. The project offers a complete framework for log-based anomaly detection, encompassing log collection, log parsing (often with a dependency on the LogParser project), feature extraction from structured log events, and the application of anomaly detection models. Loglizer is a research-backed tool, rooted in academic publications and actively maintained, providing practical implementations for practitioners and researchers in the AIOps domain. It also provides access to labeled log datasets through the Loghub project for further research and benchmarking.
https://github.com/logpai/loglizer
Also in Log Analysis & Intelligence
logpai/Drain3
Drain3 is an online log template miner that extracts structured templates from raw log messages for enhanced observability and anomaly detection.
logpai/Log3C
Log3C is a framework that identifies impactful service system problems by analyzing system logs and KPI metrics through a process of log parsing, sequence vectorization, cascading clustering, and c...