OA: Generic cluster-based anomaly detection

[detlefarend]

Description/Motivation
Cluster-based anomaly detection is based on observing changes in single or multiple cluster properties. It is imaginable to carry out the observation of single or even multiple properties in a generic way. Relevant parameters for such an algorithm could be

• Cluster property/ies to be observed
• Threshold/s for each property (lower/upper threshold -> hysteresis to avoid oscillation)
• Anomaly to be raised

Details:

• Stages of change: gradient of 1., 2., ... n. order of the property over the time ("velocity", "acceleration", ...)
• Strategies of observing multiple properties

Task list

• 1. Conception and design
  • Stage 1: single property observation
  • Stage 2: multiple property observation
• 2. Implementation (Template: mlpro.oa.streams.tasks.anomalydetection.clusterbased.AnomalyDetectorCB)
• 3. Validation
• 4. Demonstration
• 5. Documentation

Related issues
#1139

Cross references
...