Random Similarity Isolation Forests

2025-02-26

AI Summary

Existing anomaly detection methods predominantly operate on single-modality data; when handling heterogeneous features—such as time series, images, and graphs—they typically require pre-fusion or unified feature representations, which often degrade performance. This work proposes the first end-to-end multimodal anomaly detection framework that jointly models cross-modal interactions without requiring modality alignment or explicit feature transformation. Our approach innovatively integrates isolation mechanisms with cross-modal similarity projection, yielding an ensemble model grounded in random similarity kernels, isolation tree structures, and a novel multimodal distance metric. Evaluated on 47 benchmark datasets, our method consistently outperforms five state-of-the-art approaches, demonstrating statistically significant improvements in detection accuracy. The results substantiate that joint multimodal modeling—not merely feature concatenation or late fusion—yields substantial gains in anomaly detection performance, particularly for complex, real-world heterogeneous data.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

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
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

With predictive models becoming prevalent, companies are expanding the types of data they gather. As a result, the collected datasets consist not only of simple numerical features but also more complex objects such as time series, images, or graphs. Such multi-modal data have the potential to improve performance in predictive tasks like outlier detection, where the goal is to identify objects deviating from the main data distribution. However, current outlier detection algorithms are dedicated to individual types of data. Consequently, working with mixed types of data requires either fusing multiple data-specific models or transforming all of the representations into a single format, both of which can hinder predictive performance. In this paper, we propose a multi-modal outlier detection algorithm called Random Similarity Isolation Forest. Our method combines the notions of isolation and similarity-based projection to handle datasets with mixtures of features of arbitrary data types. Experiments performed on 47 benchmark datasets demonstrate that Random Similarity Isolation Forest outperforms five state-of-the-art competitors. Our study shows that the use of multiple modalities can indeed improve the detection of anomalies and highlights the need for new outlier detection benchmarks tailored for multi-modal algorithms.

Problem

• Handles mixed data types for outlier detection
• Improves anomaly detection with multi-modal data
• Introduces Random Similarity Isolation Forest algorithm

Innovation

• Isolation and similarity-based projection
• Handles arbitrary data types
• Outperforms state-of-the-art competitors