Intelligent computational techniques of machine learning models for demand analysis and prediction. (12th January 2023)

• Record Type: Journal Article
• Authors: Sundar, G. Naveen; Xavier, K. Anushka; Narmadha, D.; Sagayam, K. Martin; Jone, A. Amir Anton; Pomplun, Marc; Dang, Hien
• Is Part Of: International journal of intelligent information and database systems. Volume 16:Number 1(2023)
• Page Start: 39
• Page End: 61
• Publication Date: 2023-01-12
• Subjects: demand prediction -- machine learning -- linear regression -- feature extraction; Database management -- Computer programs -- Periodicals; Information retrieval -- Computer programs -- Periodicals; Information storage and retrieval systems -- Computer programs -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligent agents (Computer software) -- Periodicals; 006.33
• Languages: English
• ISSNs: 1751-5858

Abstract

In the proposed model, a novel approach is introduced to discover an optimal machine learning model for food demand prediction. To create an exemplary model, we used twelve different machine learning models to analyse and interpret the historical data. Feature engineering techniques have been deployed to yield better performance. All methods were evaluated using RMSE evaluation metrics to determine the optimal model. Our methodology is one of its kind to reduce the error rate to a marginal level. The novelty of our research is that the root mean square error (RMSE) value for the demand prediction was reduced to 2.61e-16 using linear regression, thus achieving a better performance. The random forest, decision tree, and extreme gradient boosting regression also performed well, producing an RMSE value of 1.42e-9, 1.93e-15, and 4.87e-18 respectively. The predictive power of the system was 100% for R-squared metrics.