BIG MART SALES PREDICTIVE ANALYSIS USING MACHINE LEARNING

Abstract

This project aimed to improve the sales forecasting abilities of Big Mart, a popular retail chain, by creating a predictive analytics model using machine learning algorithms such as XG Boost, linear regression, and time series methods. By analysing historical sales data, the model enabled Big Mart to optimise inventory levels and reduce carrying costs, while also identifying opportunities to improve sales and profitability by adjusting pricing strategies. The use of an interactive dashboard provided decision-makers with real-time access to data on sales success metrics, facilitating data- driven choices that increased revenue. The study demonstrates how machine learning and data analytics can enhance the retail sector's growth and profitability, and its findings are expected to be of great value to retail industry decision-makers. Future research may explore additional algorithms and datasets to further improve the accuracy and effectiveness of sales forecasting. Keywords: Revenue optimization, Linear Regression, XG Boost, Data Visualization.