Improved Traffic Forecasting in Transportation Systems Using Deep Learning Algorithms with Tuning Parameters

Marwan Ramdhany Edy (Universitas Negeri Makassar)

Journal of Embedded Systems, Security and Intelligent Systems
Vol 4, No 2 (2023): November 2023

In this study, we evaluated the performance of two classification models, namely Random Forest and XGBoost, on a multi-class classification task. The evaluation results showed that both produced excellent accuracy, with XGBoost achieving an accuracy of 1.00 and Random Forest around 0.99. However, Random Forest requires special attention to improve recall in some classes. These results provide important insights in the selection of classification models that fit the needs of the task. In the context of multi-class classification tasks, model performance is highly relevant and needs to be carefully calculated.