Continued...

• To achieve both high accuracy and short time requirement, many researchers are proposing various Machine Learning (ML) models. ML models have excellent capability to learn the complex relationships between the gas concentration and input features for a specific terrain.
• In our work, we analyzed three ML models, namely k-Nearest Neighbors Regression, Random Forest Regression and Neural Network. We trained and tested the performance of these models on two public gas dispersion datasets, The Prairie Grass dataset and The Round Hill dataset.