Penerapan Decision Tree Untuk Diagnosis Heat Related Ilness

Abstract

Exposure to hot temperatures is one of the work environment conditions that can cause health problems for workers, because it has become a problem that is widely found in industrial environments and can cause various health problems that have the potential to cause work accidents. Workers who are exposed to hot temperatures continuously can cause heat rash, heat cramps, heat syncope, heat exhaustion, and heat stroke. Therefore, the decission tree classification method is implemented to diagnose heat related illness. Decision Tree Learning (DTL) is one of the Machine Learning techniques that uses hierarchical sequentially structured classification rules by recursively partitioning the training data set. The process of making this decisson tree uses rapidminer software which functions as a tool to find a Decision Tree with an open classification system. This application system has various descriptive and predictive techniques in providing insight to users so that they can make the best decisions. Hail from this study shows that only 9 variables, where bright yellow or dark concentrated urine becomes the root of the tree among 18 variables in the heat related illness dataset, are most influential in determining the classification of heat related illness. This research also shows an accuracy of 89.77% in determining the classification of heat related illness using the heat related illness dataset.