Model for Predicting Prospective Big-Mart Sales Based on Grid Search Optimization (GSO)
Anantha Murthy, B. R. Puneeth, G M Harshitha, Neha Parveen, N. Balaji, Keerthi Shetty
Abstract
Predicting sales before actual sales are critical for any retailing company, to sustain a thriving business, such as Big Mart or Mall Statistical models, for example, are examples of traditional forecasting models that are frequently employed as a strategy for forecasting future sales; however, these techniques take significantly longer to estimate sales and are incapable of dealing with non-linear data. As a result, both nonlinear and linear data are dealt with using Machine Learning (ML) methodologies. ML techniques can also be used to process vast amounts of data efficiently, such as the Big Mart dataset, which has a vast amount of client data and data item attributes. A store is looking for a model that can forecast precise sales so that may anticipate consumer demand and update sale stocks ahead of time. In this study, we offer a prediction model for estimating a company's sales, such as Big. We provide a Grid Search Optimization (GSO) approach in this research for optimizing parameters and selecting the optimal tuning hyperparameters for projecting future sales of a retailing firm like Big Mart, and we discovered that our model beats others.