Analytical Methods of Machine Learning Model for E-Commerce Sales Analysis and Prediction

In the commercial market, E-commerce sales show a significant trend and have attracted many consumers. Ecommerce sales forecasting has a significant role in an organization's growth and aids in improved operation. Many studies have been conducted in the past using statistical, fundamental, and data mining techniques for better analysis and prediction of sales. However, the current scenario calls for a better study that combines the available information to propose different machine-learning techniques. The sole motive of the study is to analyze and determine different machine learning models to predict accurate results. The research observed that the Extreme Gradient Boosting model outperformed all other models and brought a good result. It produced an RMSE value of 0.0004 and Explained Variance score of 0.99. Decision Tree algorithm also shows an exemplary result.

I. Introduction

E-commerce is expanding in today’s financial world [1]. It made a paradigm shift that affected both marketers and customers. The COVID-19 crisis also hastened the growth of online shopping [2]. Customers can now purchase a variety of goods from their own residence, and businesses can continue to run despite the restrictions. Moreover, the world is evolving fast, and the business sectors are on edge with technology to meet market demands. The corporate industries are striving to satisfy customer requirements and simultaneously procure profit from their investment. Therefore, comprehending the future trend plays an important role. Sales prediction aids in estimating the upcoming sales of the firm in advance [3]. It helps in effective decision making and appropriate resource allocation. Thus, it is ideal for enhancing revenue and promoting the organization’s growth.