A Comprehensive Analysis of Regression Algorithms and Visualizations for Optimizing BigMart Sales Prediction

Accurate sales forecasting is a key area for retailers to streamline inventory management and create effective business strategies. Existing methods for predictions are not capable of reflecting the intricacy of consumer purchasing behavior, so machine learning becomes an effective option. This research investigates several regression models to predict BigMart sales. The models are assessed based on performance metrics like R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score, MAE and RMSE. Among all the models, XGBoost stands out with the highest accuracy, achieving an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.72, MAE of 698.4, and RMSE of 945.6. The outcomes show the effectiveness of XGBoost in enhancing the accuracy of sales predictions which can significantly aid retailers in making informed, data-driven decisions to improve operational efficiency and customer satisfaction