PREDICTIVE ANALYSIS FOR BIG MART SALES USING MACHINE LEARNING ALGORITHMS

Abstract

Any organization’s ability to forecast future sales is critical. Accurate forecasting of future sales helps the company improve and
manage business strategies while also gaining a thorough understanding of the economy. Standard sales forecasts assist businesses in analysing previous scenarios and applying customer purchase inferences to identify shortfalls and weaknesses before budgeting and planning for the coming year. A thorough understanding of previous opportunities allows you to anticipate future market demands and increase your chances of success. Currently, retailers like big mart rely on the traditional method of tracking sales volume items or products to forecast future customer demand and update their inventory management systems accordingly.
Incorporating Machine Learning methods to solve this can prove to be more efficient and effective. Accurate forecasting can help
businesses increase profits by a significant margin. All customer information and specific item information is stored in data warehouses, and anomalies and trends are discovered by mining the data. This mined data for retailers and companies can be used
to forecast the sales volume of various products that can potentially be purchased by the customer, which in turn helps them
stock up their inventory accordingly. Machine learning algorithms such as Linear regression, polynomial regression, Xgboost,
and Ridge regression are examples of techniques that can be used to generate predictive models for businesses with better
accuracy results. It was discovered that these models outperform traditional and existing models.