Saved in:
Bibliographic Details
Main Authors: Zi, Yun, Gong, Ming, Xue, Zhihao, Zou, Yujun, Qi, Nia, Deng, Yingnan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.09401
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912535824826368
author Zi, Yun
Gong, Ming
Xue, Zhihao
Zou, Yujun
Qi, Nia
Deng, Yingnan
author_facet Zi, Yun
Gong, Ming
Xue, Zhihao
Zou, Yujun
Qi, Nia
Deng, Yingnan
contents This study proposes an unsupervised anomaly detection method for distributed backend service systems, addressing practical challenges such as complex structural dependencies, diverse behavioral evolution, and the absence of labeled data. The method constructs a dynamic graph based on service invocation relationships and applies graph convolution to extract high-order structural representations from multi-hop topologies. A Transformer is used to model the temporal behavior of each node, capturing long-term dependencies and local fluctuations. During the feature fusion stage, a learnable joint embedding mechanism integrates structural and behavioral representations into a unified anomaly vector. A nonlinear mapping is then applied to compute anomaly scores, enabling an end-to-end detection process without supervision. Experiments on real-world cloud monitoring data include sensitivity analyses across different graph depths, sequence lengths, and data perturbations. Results show that the proposed method outperforms existing models on several key metrics, demonstrating stronger expressiveness and stability in capturing anomaly propagation paths and modeling dynamic behavior sequences, with high potential for practical deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2508_09401
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Graph Neural Network and Transformer Integration for Unsupervised System Anomaly Discovery
Zi, Yun
Gong, Ming
Xue, Zhihao
Zou, Yujun
Qi, Nia
Deng, Yingnan
Machine Learning
This study proposes an unsupervised anomaly detection method for distributed backend service systems, addressing practical challenges such as complex structural dependencies, diverse behavioral evolution, and the absence of labeled data. The method constructs a dynamic graph based on service invocation relationships and applies graph convolution to extract high-order structural representations from multi-hop topologies. A Transformer is used to model the temporal behavior of each node, capturing long-term dependencies and local fluctuations. During the feature fusion stage, a learnable joint embedding mechanism integrates structural and behavioral representations into a unified anomaly vector. A nonlinear mapping is then applied to compute anomaly scores, enabling an end-to-end detection process without supervision. Experiments on real-world cloud monitoring data include sensitivity analyses across different graph depths, sequence lengths, and data perturbations. Results show that the proposed method outperforms existing models on several key metrics, demonstrating stronger expressiveness and stability in capturing anomaly propagation paths and modeling dynamic behavior sequences, with high potential for practical deployment.
title Graph Neural Network and Transformer Integration for Unsupervised System Anomaly Discovery
topic Machine Learning
url https://arxiv.org/abs/2508.09401