This paper investigates the integration of big data analytics and machine learning techniques to enhance decision-making processes in organizational settings. Specifically, we apply supervised learning algorithms to a publicly available retail dataset containing over 50000 transaction records. Our methodology involves feature selection, model tuning, and comparative analysis across decision trees, random forests, and support vector machines. Experimental results demonstrate that the random forest classifier achieves the highest accuracy of 87.3%, significantly outperforming baseline models. These findings underscore the practical value of data-driven approaches in optimizing business strategies. The study contributes to the field by providing empirical evidence on the efficacy of machine learning for actionable insights in big data contexts.