Extended Isolation Forest

Abstract

We present an extension to the model-free anomaly detection algorithm, Isolation Forest. This extension, named Extended Isolation Forest (EIF), resolves issues with assignment of anomaly score to given data points. We motivate the problem using heat maps for anomaly scores. These maps suffer from artifacts generated by the criteria for branching operation of the binary tree. We explain this problem in detail and demonstrate the mechanism by which it occurs visually. We then propose two different approaches for improving the situation. First we propose transforming the data randomly before creation of each tree, which results in averaging out the bias. Second, which is the preferred way, is to allow the slicing of the data to use hyperplanes with random slopes. This approach results in remedying the artifact seen in the anomaly score heat maps. We show that the robustness of the algorithm is much improved using this method by looking at the variance of scores of data points distributed along constant level sets. We report AUROC and AUPRC for our synthetic datasets, along with real-world benchmark datasets. We find no appreciable difference in the rate of convergence nor in computation time between the standard Isolation Forest and EIF.

arXiv comment12 pages; 21 figures, Published. Open source code in https://github.com/sahandha/eif

References in corpus (1)

• A Meta-Analysis of the Anomaly Detection Problem

Cited by in corpus (27)

• Deep Isolation Forest for Anomaly Detection
• Anomaly Detection in Dynamic Graphs: A Comprehensive Survey
• Expert enhanced dynamic time warping based anomaly detection
• Anomaly Detection Based on Isolation Mechanisms: A Survey
• 5G Network Slice Isolation
• RO-MAP: Real-Time Multi-Object Mapping with Neural Radiance Fields
• Average Localised Proximity: A new data descriptor with good default one-class classification performance
• Prospects for a survey of the Galactic plane with the Cherenkov Telescope Array
• Interpretable Anomaly Detection with DIFFI: Depth-based Isolation Forest Feature Importance
• Detecting Backdoors in Neural Networks Using Novel Feature-Based Anomaly Detection
• Dens-PU: PU Learning with Density-Based Positive Labeled Augmentation
• A Method of Rapidly Deriving Late-type Contact Binary Parameters and Its Application in the Catalina Sky Survey
• Distance approximation using Isolation Forests
• Retrieval Augmented Deep Anomaly Detection for Tabular Data
• Cyber-Physical Systems Security: A Comprehensive Review of Anomaly Detection Techniques
• Extending Isolation Forest for Anomaly Detection in Big Data via K-Means
• PIF: Anomaly detection via preference embedding
• SPINEX: Similarity-based Predictions with Explainable Neighbors Exploration for Anomaly and Outlier Detection
• xLSTMAD: A Powerful xLSTM-based Method for Anomaly Detection
• Synthetic Data Generation for Augmenting Small Samples
• Advanced simulation-based predictive modelling for solar irradiance sensor farms
• Isolation forests: looking beyond tree depth
• Revisiting randomized choices in isolation forests
• Imputing missing values with unsupervised random trees
• Hashing for Structure-based Anomaly Detection
• Critical Review for One-class Classification: recent advances and the reality behind them
• Time Series Anomaly Detection with label-free Model Selection