Related Experiment Videos
Hybrid neuro-symbolic graph explanation framework for fraud detection in financial transaction networks
Neeraj S Kumar1, Idhikash J1, Jannath Nisha O S1
1School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India.
Introduction:
This research focuses on the development of a hybrid neuro-symbolic system for fraud detection in rapidly expanding digital financial systems.
Methods:
A bipartite customer-merchant graph is constructed from the Nigerian Financial Transactions dataset, and a 3-layer Graph Attention Network (FraudGAT) is used to model structural dependencies while a symbolic reasoning layer and constrained LLM prompting generate auditor-readable explanations.
Results:
On the curated evaluation split, FraudGAT achieves AUC 0.874, AUPRC 0.771, and F1 0.609. We additionally benchmark against GCN, GraphSAGE, GIN, a graphtransformer baseline, and tabular baselines (Random Forest, XGBoost, LightGBM): tabular ensembles perform best on this sampled split, while FraudGAT remains stronger than several graph baselines and supports an interpretable detect-explain workflow.
Discussions:
Therefore, we position the contribution as an auditable neuro-symbolic graph framework and an empirical foundation for scaling to larger, richer transaction graphs.
Related Concept Videos
01:18Signal Flow Graphs
In a signal-flow graph, branches denote the system's transfer functions, while nodes represent the signals. The direction of signal flow is indicated by arrows, with the corresponding...
01:16SFG Algebra
Each node in an SFG corresponds to a variable, and the interactions between nodes are represented by branches with associated gains. When multiple branches lead into a node, the value at that node is the sum of the...