Analisis Klasifikasi Kecelakaan Lalu Lintas Lombok Timur Berdasarkan Tingkat Keparahan Korban Kecelakaan Menggunakan Metode Support Vector Machine (SVM) dan Bootstrap Aggregating (Bagging)

Abstract

A traffic accident is a road event that occurs accidentally involving a vehicle and is experienced by fellow road users which causes casualties ranging from minor injuries, serious injuries or even causing the loss of a person's life or what is called death other than that it also results in material loss in the form of victim's property. As a result of the occurrence of traffic accidents, serious handling is needed considering that from year to year the number of traffic accidents has increased and resulted in many casualties and considerable losses. This study aims to determine the comparison of the level of accuracy of the classification results of the severity of traffic accident victims in East Lombok Regency. The method used is the Support Vector Machine (SVM) and Bootstrap Aggregating (Bagging) methods. Based on the results of the research on the East Lombok road traffic accident using the Support Vector Machine (SVM) method with under sampling and Bootstrap Aggregating (Bagging) with SMOTE, the accuracy results from these two methods show the level of accuracy support vector machine (SVM) with under sampling of 93.66% better than the Bootstrap Aggregating (Bagging) method with a SMOTE of 87.97%.