Unsupervised Graph Modeling for Anomaly Detection in Accounting Subject Relationships

TL;DR

This paper introduces an unsupervised graph neural network framework for detecting anomalies in accounting subject relationships by modeling them as a graph and analyzing structural deviations.

Contribution

It proposes a novel graph-based unsupervised method that captures both local and global anomalies in accounting data without requiring labeled examples.

Findings

Outperforms existing methods in anomaly detection accuracy.

Effectively identifies both local substructure and cross-community anomalies.

Provides traceable risk clues for accounting subject pairs.

Abstract

This paper addresses the problem of anomaly detection in accounting subject association structures, proposing a structured modeling and unsupervised discriminant framework based on graph neural networks. This framework is used to mine stable correspondences between subjects and identify structural deviations from general ledger details and voucher entries. The method first abstracts accounting subjects as graph nodes, and the co-occurrence and debit/credit correspondence of subjects in the same business record are abstracted as weighted edges. The edge weights are characterized by statistical measures such as co-occurrence frequency or amount aggregation, thus forming a period-level accounting subject association graph. In the representation learning stage, a message passing mechanism is used to fuse the node's own attributes and neighborhood context to obtain node embeddings containing structural information. In the anomaly detection stage, the rationality of subject pair connections is estimated through a relation reconstruction decoder, and edge-level anomaly scores are defined based on the degree of deviation in reconstruction probabilities. These scores are then aggregated to obtain node-level risk ranking and local anomaly localization. This framework can simultaneously capture local substructure anomalies and cross-community anomaly connections without relying on anomaly labeling, outputting traceable subject pair risk clues. Comparative experiments demonstrate more stable comprehensive discriminant capabilities and higher top-ranking accuracy.