Sales Forecasting in Retail using Machine Learning: A Case Study on Big Mart Dataset

Sales Forecasting in Retail using Machine Learning: A Case Study on Big Mart Dataset

Rishul,  Antriksh Baluni, Jatin Rajput

Department of Computer Science and Engineering (Artificial Intelligence)

IIMT College of Engineering, Greater Noida, India

Emails: rishulbosatta@gmail.com , antriksh.baluni@gmail.com, rajputjatin@gmail.com

Abstract

Stores really need to guess accurately how much they'll sell to handle what they have in stock, how they price things, and their special offers. This paper looks at using smart computer programs (machine learning) to guess these sales using old sales info from Big Mart stores. The study used different guessing methods like basic ones (Linear Regression, Ridge Regression) and a better one (Random Forest) to predict how much of each item would sell in each store. The results showed that the smarter method, Random Forest, was better at guessing and working with new data. The paper also talks about getting the data ready, making new useful information from it, and checking how good the guesses were, to give a full picture of how to predict sales in stores.

Keywords—Machine Learning, Sales Forecasting, Retail Analytics, Big Mart Dataset, Regression Models, Data Preprocessing