Agentic Anomaly Visual Analytics

Authors

Muqi Zhang (University of Konstanz), Chunpo Wu (University Konstanz), Furkan Erdi (University of Konstanz), Manuel Schmidt (University of Konstanz), Daniel A Keim (University of Konstanz), Julius Rauscher (University of Konstanz)

Abstract

We present AAVA, a web-based visual analytics tool combining a 3D activity overview, a 2D incident view, and intervention-rule evaluation. AAVA reconstructs a 192-step relay across 18 agents, uncovers two earlier incidents with the same fingerprint, and identifies one publication gate that blocks all three without affecting 105 normal posts.