Internet of Things and Graph Neural Network-Empowered Economic Effect Evaluation and Risk Control Optimization for Green Finance
Green finance promotes sustainable economic development, but evaluating its economic effectiveness and controlling financial risks remain challenging because of heterogeneous Internet of Things (IoT) data and complex relationships among financial entities. This study proposes an IoT- and Graph Neural Network-enabled Intelligent Water Drop Deep Graph Convolutional Network (IntWD-DGCNet) for economic effect evaluation and risk control optimization. A Green Finance Economic Dataset containing 16,000 records with 36 financial, environmental, investment, sustainability, and risk features is utilized. Min–max normalization and K-Nearest Neighbor imputation improve data quality, while a deep autoencoder extracts representative features. Intelligent Water Drop optimization enhances parameter tuning and graph learning, and DGCNet captures complex interconnections among entities. Experimental results show that the proposed model achieves 98.8% accuracy, 98.2% precision, 98.5% recall, and 98.7% F1-score, with statistical significance (t = 3.34, p = 0.009), demonstrating effective economic evaluation.
Authors
- Luyan Xie (ORCID: https://orcid.org/0009-0002-3809-7891)
Institutions
- Taizhou Vocational College of Science and Technology (CN)
Publication Details
- Journal
- International Journal of e-Collaboration
- Published
- 2026-11-19
- DOI
- https://doi.org/10.4018/ijec.428022
- Primary Topic
- Sustainable Finance and Green Bonds
- Type
- article
- Field-Weighted Citation Impact
- 0.00