Sales Projection by using XGBoost, Ridge Regression, Polynomial Regression & Linear Regression Algorithms in Machine Learning

Abstract

The BigMarts, the supermarket-run shopping malls, currently track sales data for each individual item to forecast possible consumer demand and revise inventory control. By mining the data store of the data warehouse, anomalies and broad trends are frequently found. The generated data can be utilised by retailers like BigMart to predict future sales volume using a variety of machine learning approaches. For projecting the sales of a company like BigMart, a predictive model was created utilising the XGBoost, Linear regression, Polynomial regression, and Ridge regression techniques. It is found that the model beats other models. To adapt the business model to anticipated outcomes, the sales estimate is based on BigMart sales for various stores. Through various machine learning techniques, the generated data may subsequently be utilised to forecast possible sales volumes for stores like BigMart. Price, outlet, and outlet location all Identifier’s included in the estimated cost of the proposed system. Many networks make use of different machine learning techniques, including linear regression, using decision tree algorithms and an XGBoost regressor, BigMart sales may be accurately predicted. Finally, hyperparameter tweaking is employed to assist in selecting pertinent hyperparameters that enhance the algorithm and yield the maximum accuracy.