Skripsi
PENGEMBANGAN MODEL TINGKAT KEPARAHAN KECELAKAAN LAUT MENGGUNAKAN METODE KLASIFIKASI MACHINE LEARNING
Maritime transportation plays an important role in international trade, more than 50,000 merchant ships are involved every day and this causes a large opportunity for maritime accidents. Therefore, to see more about the patterns and factors that influence the accidents that occur, it is necessary to classify the severity level in order to reduce and prevent the risk of more serious accidents. In the severity level classification process, machine learning methods will be used, in which the model will be developed so that the resulting predictions are more optimal. From the results of this analysis, it was found that the type of accident, type of ship, and factors contributed significantly to the severity of the accident, around 50.3% of the human role was the main causal factor in the occurrence of the accident. Based on the results of the comparison of two machine learning models, one of the best models was obtained, namely LGBM (Light Gradient Boosting Machine Classifier) by hyperparameterizing the model.
No other version available