A Security Monitoring and Multi-source Anomaly Detection Method for New Distributed Systems Oriented to Economic Data Asset Circulation

To address the difficulty of unified security risk monitoring caused by the coupling of on-chain auditing, microservice invocation, distributed databases, and network communication states during economic data asset circulation, this paper proposes a multi-source spatio-temporal graph anomaly detection method for new distributed systems, namely MSTG-AD. The method aligns five monitoring sources—transaction behavior, on-chain auditing, microservice invocation, database metrics, and network traffic—with a six-entity relation schema covering users, data assets, contracts, services, databases, and network nodes. Rather than claiming novelty for the individual temporal, Transformer, graph-attention, or spectral operators, the proposed contribution lies in their cross-layer semantic coupling: multi-window temporal encoding captures periodic changes and abrupt fluctuations, call-chain/database encoding characterizes service and node states, an asset-context-aware gate prioritizes anomaly evidence according to value and exposure context, and a relation-constrained decoder traces temporally consistent root-cause paths along the economic-data-asset chain. On a simulated economic data asset trading platform and a multi-source fusion test set, MSTG-AD achieves a Precision of 95.1%, a Recall of 94.0%, an F1-score of 94.5%, and an AUC of 0.973, with an average detection latency of 38.6 ms. In the root cause localization task, AC@1, AC@3, and AC@5 reach 0.91, 0.96, and 0.98, respectively. The results demonstrate the effectiveness of full-link multi-source modeling for anomaly identification, risk monitoring, and explainable root-cause localization.

Authors

Institutions

Publication Details

Journal
ICST Transactions on Scalable Information Systems
Published
2026-11-18
DOI
https://doi.org/10.4108/eetsis.13812
Primary Topic
Software System Performance and Reliability
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Security Monitoring and Multi-source Anomaly Detection Method for New Distributed Systems Oriented to Economic Data Asset Circulation

Chenfei Zhuang, Xingchao Ni, Jiajia Cao
1 links
ICST Transactions on Scalable Information Systems
Software System Performance and Reliability
article

A Security Monitoring and Multi-source Anomaly Detection Method for New Distributed Systems Oriented to Economic Data Asset Circulation

Chenfei Zhuang, Xingchao Ni, Jiajia Cao
article en

Abstract

To address the difficulty of unified security risk monitoring caused by the coupling of on-chain auditing, microservice invocation, distributed databases, and network communication states during economic data asset circulation, this paper proposes a multi-source spatio-temporal graph anomaly detection method for new distributed systems, namely MSTG-AD. The method aligns five monitoring sources—transaction behavior, on-chain auditing, microservice invocation, database metrics, and network traffic—with a six-entity relation schema covering users, data assets, contracts, services, databases, and network nodes. Rather than claiming novelty for the individual temporal, Transformer, graph-attention, or spectral operators, the proposed contribution lies in their cross-layer semantic coupling: multi-window temporal encoding captures periodic changes and abrupt fluctuations, call-chain/database encoding characterizes service and node states, an asset-context-aware gate prioritizes anomaly evidence according to value and exposure context, and a relation-constrained decoder traces temporally consistent root-cause paths along the economic-data-asset chain. On a simulated economic data asset trading platform and a multi-source fusion test set, MSTG-AD achieves a Precision of 95.1%, a Recall of 94.0%, an F1-score of 94.5%, and an AUC of 0.973, with an average detection latency of 38.6 ms. In the root cause localization task, AC@1, AC@3, and AC@5 reach 0.91, 0.96, and 0.98, respectively. The results demonstrate the effectiveness of full-link multi-source modeling for anomaly identification, risk monitoring, and explainable root-cause localization.

ICST Transactions on Scalable Information SystemsVol. 13(8)
Suzhou University of Technology (CN), Nantong Institute of Technology (CN)
Software System Performance and Reliability
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon