A Comprehensive Survey on Sales Forecasting Models Using Machine Learning Algorithms

A Comprehensive Survey on Sales Forecasting Models Using Machine Learning Algorithms

연구 분야: Artificial Intelligence

학회: 2022 Fourth International Conference on Emerging Research in Electronics, Computer Science and Technology (ICERECT)

Sales Forecasting is a most commonly used in marketing. Nowadays a large number of companies are using this technique to manufacture their product. In this we are going to study about the usage of different Machine Learning Models and Techniques used for sales prediction. The overall study of models and techniques is to increase the efficiency of future sales prediction. Nowadays for any product there are lakhs of reviews were generated by the users on different products in the market. Which confuses customers to make decision whether to buy product or not. And for a specific company to study overall reviews is hard to make product manufacture. This study mainly deals with arranging the opinions of different customers and different kinds of techniques used in sales forecasting. The present work uses mainly four machine learning algorithms namely Support Vector Machine (SVM), Decision Tree (DT), Linear Regression, Random Forest, and K-Nearest Neighbors, K-means Clustering, Logistic Regression for classifying reviews. The forecasting accuracy of each algorithm is evaluated with the Root Mean Square Error (RMSE). The study found that Random Forest is the best model because it had lowest Root Mean Square Error (RMSE) compared to other model and for classifying reviews the Logistic Regression is giving accurate result.