RFOD: Random Forest-based Outlier Detection for Tabular Data

2025-10-09

AI Summary

To address semantic loss, poor interpretability, and modeling challenges in anomaly detection for mixed-type tabular data, this paper proposes RFOD—a Random Forest-based Outlier Detection framework. RFOD reformulates anomaly detection as a feature-level conditional reconstruction task, employing independent random forests to model each feature without imposing global distributional assumptions. It introduces two key innovations: Adjusted Gower Distance (AGD) for heterogeneous feature similarity computation, and Uncertainty-Weighted Averaging (UWA) for aggregating cell-level reconstruction errors—enabling fine-grained, interpretable, cell-wise anomaly scoring. Crucially, RFOD inherently preserves categorical semantics without manual encoding. Evaluated on 15 real-world heterogeneous datasets, RFOD consistently outperforms state-of-the-art methods across detection accuracy, robustness, scalability, and interpretability—demonstrating particular efficacy for high-stakes tabular anomaly identification.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Feature Construction/Reformulation
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Outlier detection in tabular data is crucial for safeguarding data integrity in high-stakes domains such as cybersecurity, financial fraud detection, and healthcare, where anomalies can cause serious operational and economic impacts. Despite advances in both data mining and deep learning, many existing methods struggle with mixed-type tabular data, often relying on encoding schemes that lose important semantic information. Moreover, they frequently lack interpretability, offering little insight into which specific values cause anomalies. To overcome these challenges, we introduce extsf{ extbf{RFOD}}, a novel extsf{ extbf{R}}andom extsf{ extbf{F}}orest-based extsf{ extbf{O}}utlier extsf{ extbf{D}}etection framework tailored for tabular data. Rather than modeling a global joint distribution, extsf{RFOD} reframes anomaly detection as a feature-wise conditional reconstruction problem, training dedicated random forests for each feature conditioned on the others. This design robustly handles heterogeneous data types while preserving the semantic integrity of categorical features. To further enable precise and interpretable detection, extsf{RFOD} combines Adjusted Gower's Distance (AGD) for cell-level scoring, which adapts to skewed numerical data and accounts for categorical confidence, with Uncertainty-Weighted Averaging (UWA) to aggregate cell-level scores into robust row-level anomaly scores. Extensive experiments on 15 real-world datasets demonstrate that extsf{RFOD} consistently outperforms state-of-the-art baselines in detection accuracy while offering superior robustness, scalability, and interpretability for mixed-type tabular data.

Problem

• Detects outliers in mixed-type tabular data
• Handles heterogeneous data without losing semantic information
• Provides interpretable anomaly detection with cell-level insights

Innovation

• Random Forest-based outlier detection for tabular data
• Feature-wise conditional reconstruction using dedicated forests
• Cell-level scoring with Adjusted Gower's Distance and Uncertainty-Weighted Averaging