<p>Financial inclusion is vital for promoting economic development and alleviating poverty, especially in developing economies where many individuals and businesses are underserved. Traditional methods for assessing creditworthiness often rely on static models that fail to convey the fluidity of economic actions. This study suggests a Multi-Component Attention Graph Convolutional Network (MCAGCN) to overcome these drawbacks to enhance both credit assessment accuracy and service personalization. The approach includes several key steps: utilizing diverse datasets to train the MCAGCN, which captures complex relationships among individuals and financial activities. The MCAGCN framework is implemented in Python, and its performance is compared with existing methods. Results indicate that this novel approach significantly improves credit assessment accuracy and accessibility, facilitating better financial inclusion for marginalized populations and promoting their engagement in the formal economy.</p>