An Improved Time Series Anomaly Detection by Applying Structural Similarity

2025-09-24

AI Summary

Existing unsupervised time-series anomaly detection methods based on reconstruction rely solely on point-wise distances, neglecting structural characteristics such as trends, seasonality, and shape patterns, thereby failing to identify pattern-level anomalies. This paper proposes StrAD, the first approach to explicitly incorporate multi-scale structural similarity—encompassing trend, periodicity, and local morphology—into the reconstruction objective. StrAD introduces a plug-and-play structure-aware mechanism that jointly enforces global fluctuation consistency and local feature fidelity. Within a standard unsupervised reconstruction framework, it leverages structural priors to guide model training, significantly enhancing joint detection of both point-level and pattern-level anomalies. Extensive experiments on five real-world benchmark datasets demonstrate that StrAD consistently outperforms state-of-the-art reconstruction-based methods, achieving an average AUC improvement of 4.2%.

Technology Category

• Machine Learning: Structured Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Effective anomaly detection in time series is pivotal for modern industrial applications and financial systems. Due to the scarcity of anomaly labels and the high cost of manual labeling, reconstruction-based unsupervised approaches have garnered considerable attention. However, accurate anomaly detection remains an unsettled challenge, since the optimization objectives of reconstruction-based methods merely rely on point-by-point distance measures, ignoring the potential structural characteristics of time series and thus failing to tackle complex pattern-wise anomalies. In this paper, we propose StrAD, a novel structure-enhanced anomaly detection approach to enrich the optimization objective by incorporating structural information hidden in the time series and steering the data reconstruction procedure to better capture such structural features. StrAD accommodates the trend, seasonality, and shape in the optimization objective of the reconstruction model to learn latent structural characteristics and capture the intrinsic pattern variation of time series. The proposed structure-aware optimization objective mechanism can assure the alignment between the original data and the reconstructed data in terms of structural features, thereby keeping consistency in global fluctuation and local characteristics. The mechanism is pluggable and applicable to any reconstruction-based methods, enhancing the model sensitivity to both point-wise anomalies and pattern-wise anomalies. Experimental results show that StrAD improves the performance of state-of-the-art reconstruction-based models across five real-world anomaly detection datasets.

Problem

• Detecting anomalies in time series without labeled data
• Overcoming limitations of point-by-point distance measures
• Capturing structural patterns like trends and seasonality

Innovation

• Incorporates structural similarity into reconstruction objectives
• Learns trend, seasonality and shape characteristics jointly
• Pluggable mechanism compatible with existing reconstruction models