On PaySim, graph and autoencoder signals improve fraud ranking only among mid-confidence cases, and a bounded LLM agent underperforms direct classifier thresholding.
Graph neural networks for financial fraud detec- tion: A review.Frontiers of Computer Science, 2025
1 Pith paper cite this work, alongside 87 external citations. Polarity classification is still indexing.
1
Pith paper citing it
87
external citations · OpenAlex
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Toward Auditable Fraud Detection: Combining Graph Features, Model Explanations, and Agentic Case Investigation
On PaySim, graph and autoencoder signals improve fraud ranking only among mid-confidence cases, and a bounded LLM agent underperforms direct classifier thresholding.