Explainable Anomaly Detection for Industrial IoT Data Streams
This work addresses the challenge of explainable and adaptive maintenance decisions for industrial IoT systems, though it appears incremental as it builds on existing data stream mining and anomaly detection methods.
The paper tackles the problem of real-time anomaly detection in industrial IoT data streams where labels are often unavailable, by presenting a collaborative framework that integrates unsupervised anomaly detection with human-in-the-loop learning, and provides initial results for fault detection in a Jacquard loom unit.
Industrial maintenance is being transformed by the Internet of Things and edge computing, generating continuous data streams that demand real-time, adaptive decision-making under limited computational resources. While data stream mining (DSM) addresses this challenge, most methods assume fully supervised settings, yet in practice, ground-truth labels are often delayed or unavailable. This paper presents a collaborative DSM framework that integrates unsupervised anomaly detection with interactive, human-in-the-loop learning to support maintenance decisions. We employ an online Isolation Forest and enhance interpretability using incremental Partial Dependence Plots and a feature importance score, derived from deviations of Individual Conditional Expectation curves from a fading average, enabling users to dynamically reassess feature relevance and adjust anomaly thresholds. We describe the real-time implementation and provide initial results for fault detection in a Jacquard loom unit. Ongoing work targets continuous monitoring to predict and explain imminent bearing failures.