PANDA: Human-in-the-Loop Anomaly Detection and Explanation

Grégory Smits (1, 2) , Marie-Jeanne Lesot (3) , Véronne Yepmo (4, 1, 2) , Olivier Pivert (1, 2)

1 IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires
2 SHAMAN - A Symbolic and Human-centric view of dAta MANagement
3 LFI - Learning, Fuzzy and Intelligent systems
4 UR - Université de Rennes

The paper addresses the tasks of anomaly detection and explanation simultaneously, in the human-in-the-loop paradigm integrating the end-user expertise: it first proposes to exploit two complementary data representations to identify anomalies, namely the description induced by the raw features and the description induced by a user-defined vocabulary. These representations respectively lead to identify so-called data-driven and knowledge-driven anomalies. The paper then proposes to confront these two sets of instances so as to improve the detection step and to dispose of tools towards anomaly explanations. It distinguishes and discusses three cases, underlining how the two description spaces can benefit from one another, in terms of accuracy and interpretability.

Mots clés: outlier detection; outlier explanation; XAI; human-in-theloop; fuzzy vocabulary; linguistic description

Domaines: Recherche d'information [cs.IR]; Base de données [cs.DB]

Grégory Smits, Marie-Jeanne Lesot, Véronne Yepmo, Olivier Pivert. PANDA: Human-in-the-Loop Anomaly Detection and Explanation. IPMU 2022 - Information Processing and Management of Uncertainty in Knowledge-Based Systems, Jul 2022, Milan, Italy. ⟨hal-03696295⟩