Accurate temperature forecasting plays a vitalrole in supporting data-driven decision-making across varioussectors including energy planning, environmental comfortmanagement, and public safety. This study investigates theapplication of Random Forest Regression in a limited dataset forshort-term air temperature prediction. Specifically, RandomForest is designed to forecast hourly temperature values for thenext seven days using a minimal dataset upon a single day ofmeteorological observations. The observation includestemperature, humidity, atmospheric pressure, and wind speed.Despite of the limited temporal scope of dataset, Random Forestmodel demonstrates a notable ability to simulate realistictemperature patterns. The results reveal that Random Forest iscapable of handling heterogeneous input features and deliveringaccurate predictions under normal environmental conditioneven with constrained data availability. Accordingly, thisresearch yields the convergence of MAE (Mean Absolute Error)is 0.102 and RMSE (Root Mean Square Error) value is 0.136.The findings underscore the potential of Random Forest forshort-term temperature forecasting in data-limited scenarios.