Paper Title
Predicting SDG Index and Goal Indicators Using Machine Learning: A Data-Driven Approach for Sustainable Development Analysis

Abstract
Achieving the United Nations Sustainable Development Goals (SDGs) by 2030 necessitates a robust understanding of the intricate relationships among economic, social, environmental, and geographic factors. This study introduces an advanced machine learning framework leveraging XGBoost with Recursive Feature Elimination (RFE) and hyperparameter tuning to predict SDG Index Scores and analyze regional disparities. By integrating multidimensional datasets and exploring geographic aggregation techniques, the research reveals that regional-level analysis significantly enhances predictive accuracy. The results demonstrate the superiority of XGBoost in capturing spatial dynamics and improving model performance, with an R² of 99.04% in the best configuration. This study underscores the critical role of geographic attributes in understanding SDG trajectories and provides actionable insights for policymakers. By offering a scalable, data-driven methodology, it advances the analytical tools available for sustainable development analysis and highlights the need for targeted interventions in underperforming regions. Keywords - Sustainable Development Goals (SDGs), Machine Learning, XGBoost, Recursive Feature Elimination (RFE), Predictive Modeling, SDG Index Score.