Publication notice: SYNASC 2023
Conference: SYNASC 2023: 25th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing
We are delighted to announce the recent publication of research by our partner University Babeș-Bolyai done during the DataSeer project, at SYNASC 2023. The paper compares the statistical relevance of neural networks and decision tree methods in forecasting popular beverage consumption. By analyzing a significant sample of millions of observations over multiple years, various error metrics such as MALE, MSLE, RMSLE, GMRE, and ERMSLE were computed. The results aim to identify the most accurate model for predicting product consumption and supporting reliable business decision-making processes.