Function Based Isolation Forest (FuBIF): A Unifying Framework for Interpretable Isolation-Based Anomaly Detection

2025-11-08

AI Summary

To address the rigid partitioning strategy and limited interpretability of Isolation Forest (IF) in anomaly detection, this paper proposes Functional Isolation Forest (FuBIF), the first IF variant that replaces axis-aligned splits with real-valued functional splits, enabling more flexible and adaptive data partitioning. We further introduce FuBIFFI, a unified feature importance attribution method applicable to any FuBIF model, enhancing post-hoc explainability. Extensive experiments on standard benchmark datasets demonstrate that FuBIF significantly outperforms classical IF and state-of-the-art anomaly detection methods in detection accuracy. The open-source implementation ensures full reproducibility. This work unifies and extends the theoretical framework of isolation forests, simultaneously improving detection performance, model transparency, and debuggability.

Technology Category

• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Ensemble Methods
• Computer Vision: Interpretability, Explainability, and Transparency

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Anomaly Detection (AD) is evolving through algorithms capable of identifying outliers in complex datasets. The Isolation Forest (IF), a pivotal AD technique, exhibits adaptability limitations and biases. This paper introduces the Function-based Isolation Forest (FuBIF), a generalization of IF that enables the use of real-valued functions for dataset branching, significantly enhancing the flexibility of evaluation tree construction. Complementing this, the FuBIF Feature Importance (FuBIFFI) algorithm extends the interpretability in IF-based approaches by providing feature importance scores across possible FuBIF models. This paper details the operational framework of FuBIF, evaluates its performance against established methods, and explores its theoretical contributions. An open-source implementation is provided to encourage further research and ensure reproducibility.

Problem

• Generalizes Isolation Forest to use real-valued functions for dataset branching
• Enhances flexibility and reduces biases in anomaly detection tree construction
• Extends interpretability by providing feature importance scores across models

Innovation

• FuBIF generalizes Isolation Forest with real-valued functions
• FuBIFFI algorithm provides feature importance scores
• Framework enhances flexibility in anomaly detection trees