Klasifikasi Tingkat Keparahan Kecelakaan Lalu Lintas Menggunakan Random Forest Classifier

Abstract

Traffic accidents are a common problem that often occurs. Many factors cause and determine the severity of traffic accidents. These factors can include road conditions, weather, light conditions, driver age, and the cause of the accident. In this study, researchers will try to apply the Random Forest method to classify the severity of traffic accidents. The Random Forest method was chosen because of its excellent ability to handle high-dimensional data and tolerance for overfitting. The dataset used in this research was taken from Kaggle, consisting of 12316 records and 32 features covering various attributes related to traffic accidents. Before applying random forest, it is necessary to carry out a preprocessing stage on the dataset to remove irrelevant features, fill in empty values ??and divide the data into training and testing data. The final results of this research show that Random Forest can produce a good level of in classifying the severity of traffic accidents with 92% accuracy. This shows the potential of this method as a useful tool in the analysis and prediction of traffic accidents. Therefore, this research makes a significant contribution to efforts to improve road safety.

Keywords: Random Forest Classifier, Traffic accident, Machine learning, Data Classification, Supervised Learning