Innovative Fraud Detection in Supply Chain Finance using Graph Attention Networks
- 5 February 2025
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1381-1386
Abstract
Supply Chain Finance (SCF) is essential to streamlining supply chain capital activities and lowering management costs. Big Data, the IoT, AI, and Blockchain have helped SCF make better investment decisions and create personalized risk pricing models. Fast development of these technologies has led to a more nuanced distribution of unique financial fraud patterns among usual patterns and a huge increase of SCF data. Our method involves training, feature extraction, and model preparation. More data preparation was done after bias discovery and feature selection to prepare the data for machine learning models. Linear Discriminant Analysis (LDA) in feature extraction identified the descriptions' hidden subjects. Our models were trained using HGAT, or Hierarchical Graph Attention Network. Graph Neural Networks (GNN) and traditional HGAT were typically outperformed by the suggested model at 94.34% accuracy. This model can improve supply chain finance decision-making and risk management since our method detects subtle SCF financial fraud tendencies more accurately.Keywords
This publication has 16 references indexed in Scilit:
- Optimizing E-Commerce Fraud Detection with BiGRU and Capsule Network Architectures2024
- Online Transaction Fraud Detection: Exploring the Hybrid SSA-TCN-BiGRU Approach2024
- Supply chain fraud prediction with machine learning and artificial intelligenceInternational Journal of Production Research, 2024
- Temporal Threat Recognition in Supply Chains: Integrating Hidden Markov Models for Proactive Security with AI-Driven Automated Threat Hunting2024
- Towards a Framework for Supply Chain Financing for Order-Level Risk Prediction: An Innovative Stacked A-GRU Based Technique2024
- Intelligent System for Fraud Detection in Online Banking using Improved Particle Swarm Optimization and Support Vector Machine2023
- Retracted: Application of Artificial Intelligence in the Supply Chain Finance2023
- Using Machine Learning Approach to Evaluate the Excessive Financialization Risks of Trading EnterprisesComputational Economics, 2021
- Leveraging deep learning with LDA-based text analytics to detect automobile insurance fraudDecision Support Systems, 2018
- A credit risk assessment model based on SVM for small and medium enterprises in supply chain financeFinancial Innovation, 2015