Comparative Analysis of Regression Algorithms used to Predict the Sales of Big Marts
DOI:
https://doi.org/10.56536/jicet.v3i1.53Abstract
Abstract— Sales predictions or forecasting can help in analyzing the current and future sales trends of a big mart company. Based on the sales prediction or forecast, a retailer company can plan its production, marketing and promotional activities. Using several machine learning techniques, the obtained data may then be utilized to predict possible sales for retailers. This paper investigates that which machine learning regression algorithm best predicts big marts sales and which technique has the highest correlation coefficient value and the lowest values of mean absolute error (MAE), relative absolute error (RAE), root mean squared error (RMSE), and root relative squared error (RRSE). A comparative analysis of various machine learning regression algorithms such as SMO regression, simple linear regression, linear regression, additive regression, multi-layer perceptron, random forest, and M5P will be provided in this paper. After the experiments are completed, a comparison of various cross validations and splitting ratios for training and testing data will be given.