Unified Taxonomy for Multivariate Time Series Anomaly Detection using Deep Learning

2026-03-19

AI Summary

This study addresses the lack of a systematic classification framework for multivariate time series anomaly detection, which has impeded methodological comparison and evolutionary analysis. To bridge this gap, the work proposes a unified taxonomy encompassing eleven dimensions organized into three major categories: input, output, and model. This framework is the first to offer a structured, comprehensive, and extensible classification system for deep learning–based approaches in this domain. Through an extensive literature review and methodological analysis—combining qualitative synthesis with confirmatory evaluation—the validity of each dimension is rigorously established and verified. The analysis reveals a prevailing trend toward Transformer-based architectures and reconstruction- or prediction-oriented paradigms, while also laying the groundwork for future advancements in adaptive and generative methodologies.

Technology Category

Data Mining & Knowledge Management: Anomaly/Outlier DetectionMachine Learning: Multi-class/Multi-label Learning & Extreme ClassificationCognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsWeb Mining and Content Analysis: Models for Web evolutionUser Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

The topic of Multivariate Time Series Anomaly Detection (MTSAD) has grown rapidly over the past years, with a steady rise in publications and Deep Learning (DL) models becoming the dominant paradigm. To address the lack of systematization in the field, this study introduces a novel and unified taxonomy with eleven dimensions over three parts (Input, Output and Model) for the categorization of DL-based MTSAD methods. The dimensions were established in a two-fold approach. First, they derived from a comprehensive analysis of methodological studies. Second, insights from review papers were incorporated. Furthermore, the proposed taxonomy was validated using an additional set of recent publications, providing a clear overview of methodological trends in MTSAD. Results reveal a convergence toward Transformer-based and reconstruction and prediction models, setting the foundation for emerging adaptive and generative trends. Building on and complementing existing surveys, this unified taxonomy is designed to accommodate future developments, allowing for new categories or dimensions to be added as the field progresses. This work thus consolidates fragmented knowledge in the field and provides a reference point for future research in MTSAD.

Problem

• Multivariate Time Series Anomaly Detection
• Deep Learning
• Taxonomy
• Systematization
• Methodological Trends

Innovation

• unified taxonomy
• multivariate time series anomaly detection
• deep learning
• Transformer-based models
• systematic categorization