Anomaly Detection helps you detect anomalies in your system’s log using machine learning algorithms and helps you investigate and resolve abnormal activities occurring in your application faster.
How it works :
The 2 major components working behind it are Time Series Analysis and Semantic Analysis.
- Time Series Analysis: This component detects the unusual patterns in log occurrence frequency over time
- Semantic Analysis: This component identifies unusual or unexpected content patterns within the log messages
For log anomaly, logs are constantly monitored and passed through both processes and if it gets over to threshold then it will be shown as an anomaly.
The formula behind the scenes is (Total Anomaly Counts per Minute / Total Log Counts per Minute) > Threshold
For more information, you can check the documentation here: Anomaly Detection
