<![CDATA[Rain can have various impacts on human life, one of which is the impact on the world of aviation. To minimize the impact caused by rain in the world of aviation, rain forecasts are needed to facilitate operational activities at an airport, including Sultan Hasanuddin International Airport, which is one of the busiest airports in Indonesia. In forecasting rain, generally the data used is data that influences the formation of rain, such as data related to the amount of water vapor (precipitable water) and wind (relative vorticity and divergence). Even though the data used in forecasting rain is correlated with the formation of rain, there is the potential for poor forecast accuracy due to non – continuous supporting data for predicted rain events because atmospheric conditions are very complex and can change rapidly. To minimize rain forecasts with poor accuracy, a method is needed that can process non – continuous data regarding future rain events well, one of which is using the Decision Tree C4.5 algorithm. Decision Tree C4.5 is a machine learning algorithm that involves selecting the best features at each step so that it has the potential to produce good forecasts. In this study, the forecast results for a year were dominated by 2590 no rain events, while the total number of rain forecasts was 330 events. Accuracy of monthly forecasts was found to range from 64.92% to 100%, where if the number of correct and incorrect forecasts for each month were combined, the forecast accuracy for a year was 84%, where this accuracy could be said to be very good.]]>
Copyrights © 2024
