Main Article Content
Sales forecasting is a fundamental assignment in retailing. In such manner, the usage of machine-learning models for sales predictive investigation were studied. The goal is to review scientific literature and distinguish if there are advantages over conventional statistical techniques. The detailed study and examination of predictive models is to improve future sales predictions are completed in this study. Millions of reviews are being created daily which makes it hard for a consumer to make a good decision on whether to shop for the product. Investigating this huge number of opinions is hard and tedious for product manufacturers. This work considers the issue of arranging reviews by their overall semantic (positive or negative) In this paper, the proposed work uses three machine learning algorithms namely Linear Regression, Decision Tree (DT) and Random Forest (RF) in sales prediction and Logistic Regression for classifying the reviews. The forecasting precision of each scenario is evaluated with the Root Mean Square Error (RMSE). The examinations found that the best model is Random Forest Algorithm, which shows greatest precision in forecasting and in future sales prediction as it had a lower mean absolute error than the other two models.
This work is licensed under a Creative Commons Attribution 4.0 International License.