Open AccessVessel Classifying and Trajectory Based on Automatic Identification System Data
Author(s) -
Natalia Damastuti,
Aulia Siti Aisjah,
Agoes A. Masroeri
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/830/1/012049
Subject(s) - automatic identification system , trajectory , identification (biology) , computer science , mean squared error , process (computing) , data mining , root mean square , value (mathematics) , artificial intelligence , pattern recognition (psychology) , statistics , mathematics , engineering , machine learning , botany , astronomy , biology , electrical engineering , operating system , physics
Nowadays, the development of the of Automatic Identification System (AIS) device has continuously increased. It was initially used to send information on the whereabouts of ships to avoid collisions, but with stored data, it is used for monitoring waters. Therefore, this study was carried out using AIS data to classify ships in Indonesian waters. Based on features such as length, width, and weight, it classified them into 9 types of vessels. The data mining process was used to characterize each type with the ensemble method. Furthermore, data processing was carried out to determine the ship’s trajectory pattern. In this study, 80% of training data was used while the rest were testing data. The results showed that an accuracy value of 99.8% was obtained with a Root Mean Square Error (RMSE) value of 0.12.