Optimization of Data Quality for Green Finance Using XGBoost Algorithm: Integration of Financial Market Regulations
Abstract
The significance of high-quality data has increased due to the growing complexity of achieving sustainable growth within the context of economic marketplace rules. Nevertheless, there is no comprehensive strategy that combines the necessary data optimization with green finance (GF) measures. This investigation offers a comprehensive technique that combines data optimization with GF measures. Using a sample of thirty companies registered on the Saudi Share Trade (SST), the investigation employs factor analysis approaches. This study develops a GF model that maintains sustainability in a sample of enterprises selected using a potent ensemble approach called XGBoost (eXtreme Gradient Boosting). An approach known as the Synthetic Minority Oversampling Technique (SMOTE) is used to resolve the database’s group unbalance. This study employs various ML methods in Python for GF. The experimental results indicate that the XGBoost method functioned better than the other ML methods. Next, this study optimizes the XGBoost technique to achieve the optimal outcome, resulting in an overall GF accuracy of 96.1%. Shareholders and governments will both benefit from the research’s conclusions. These prediction algorithms can be used by regulators to create well-balanced policy programs that promote GFs and ensure financial development. Simultaneously, the analytically profitable system features identified in this investigation will enable traders and users to easily adjust to rapid shifts in the marketplace and generate income.