Deep Context-Conditioned Anomaly Detection for Tabular Data

2025-09-10

AI Summary

To address the problem that globally rare events in large-scale tabular data are misclassified as normal within local heterogeneous contexts (e.g., distinct users), this paper proposes the first context-aware unsupervised anomaly detection framework specifically designed for tabular data. The method automatically identifies salient contextual features and models the conditional data distribution using a deep autoencoder—eschewing global distributional assumptions to precisely capture locally normal patterns. Crucially, it innovatively incorporates a context-conditioning mechanism into tabular anomaly detection, enabling end-to-end adaptive learning. Extensive experiments on multiple standard tabular benchmarks demonstrate significant improvements over existing state-of-the-art methods. Results empirically validate that explicit contextual modeling is essential for accurate anomaly detection and yields substantial performance gains.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Unsupervised & Self-Supervised Learning
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Anomaly detection is critical in domains such as cybersecurity and finance, especially when working with large-scale tabular data. Yet, unsupervised anomaly detection -- where no labeled anomalies are available -- remains a significant challenge. Although various deep learning methods have been proposed to model a dataset's joint distribution, real-world tabular data often contain heterogeneous contexts (e.g., different users), making globally rare events normal under certain contexts. Consequently, relying on a single global distribution can overlook these contextual nuances, degrading detection performance. In this paper, we present a context-conditional anomaly detection framework tailored for tabular datasets. Our approach automatically identifies context features and models the conditional data distribution using a simple deep autoencoder. Extensive experiments on multiple tabular benchmark datasets demonstrate that our method outperforms state-of-the-art approaches, underscoring the importance of context in accurately distinguishing anomalous from normal instances.

Problem

• Detecting anomalies in heterogeneous tabular data contexts
• Addressing unsupervised anomaly detection without labeled examples
• Modeling conditional distributions to capture contextual normalcy variations

Innovation

• Context-conditional anomaly detection framework
• Autoencoder models conditional data distribution
• Automatically identifies contextual features