Leveraging Ensemble Machine Learning to Evaluate Financial Inclusion and Social Equity: Insights from CatBoost and Gradient Boosting Models

For long-term economic growth, it is important to have financial inclusion and social fairness. However, it is hard to measure these things because socioeconomic data is complicated and doesn't follow a straight line. This work introduces a machine learning ensemble architecture that integrates the characteristics of the CatBoost and Gradient Boosting models to assess social equality and financial inclusion. A variety of AI-driven and digital transformation indicators are employed to evaluate the institutional and organizational factors that affect inclusion outcomes. These metrics include how well data is used, how well cloud computing is used, how fast decisions are made, how happy customers are, how innovative AI is, how well risks are managed, and how well AI follows ethical guidelines. When training and testing models, standard performance indicators include F1-score, recall, accuracy, and precision. The experimental results show that CatBoost is better at handling categorical data and less likely to overfit than Gradient Boosting on all evaluation metrics. Research comparing different levels of financial inclusion has found that higher levels of digital maturity are linked to happier customers, more innovation, and following ethical AI principles. The findings indicate that ensemble learning methodologies are proficient in evaluating social equity and financial inclusion. These procedures are easy to learn, reliable, and may be used on a large scale. This report gives important information to development agencies, banks, and members of parliament so they can set up data-driven programs that encourage growth for everyone and make it easier for everyone to get financial services in the digital age.