Analysing Social Entrepreneurship and Financial Inclusion for Women Empowerment Using an F-GNN and AXGB Hybrid Approach

Financial exclusion continues to appear a major barrier, especially for women in developing nations, restricting their access to credit, digital transactions, and entrepreneurial advancement. Conventional prediction models improperly represent the complex interconnections among financial behaviours, social entrepreneurship, and economic empowerment. A hybrid Factor Graph Neural Network (F-GNN) and Adaptive eXtreme Gradient Boosting (AXGB) model is presented to utilise graph-based learning for complex dependency modelling and boosted decision trees for improved classification accuracy. The model, developed using real-world datasets, surpasses baseline methods, attaining 97.2% accuracy, 96.5% precision, and a 98.3% ROC-AUC score, illustrating its superiority in predicting financial inclusion. Findings indicate that social entrepreneurship enhances financial knowledge, loan accessibility, and economic independence. This study enhances AI-based financial analytics, providing a scalable, high-performance solution for policymakers, financial institutions, and social businesses.