Identification of Damaged AIS Data Based on Clustering and Multi-Label Classification
Data
2023
Autorzy
Tytuł czasopisma
ISSN czasopisma
Tytuł tomu
Wydawca
Wydawnictwo Politechniki Łódzkiej
Lodz University of Technology Press
Lodz University of Technology Press
Abstrakt
Automatic Identification System (AIS) is a telecommunication system
that allows ships to communicate with each other by sending information
regarding their trajectory: location, speed, course and so on. Due to some
technical issues, some parts of the transmitted data might be damaged or
incomplete. In this paper, we propose a machine learning based approach
for detecting AIS data that requires reconstruction. The general idea of the
proposed approach, utilizing clustering and mutli-label classification algorithms,
and its performance are discussed.
Opis
Słowa kluczowe
AIS data analysis, anomaly detection, multilabel classification, automatic identification system, analiza danych AIS, wykrywanie anomalii, klasyfikacja wieloetykietowa, system automatycznej identyfikacji
Cytowanie
Szarmach M., Czarnowski I., Identification of Damaged AIS Data Based on Clustering and Multi-Label Classification. W: Progress in Polish Artificial Intelligence Research 4, Wojciechowski A. (Ed.), Lipiński P. (Ed.)., Seria: Monografie Politechniki Łódzkiej Nr. 2437, Wydawnictwo Politechniki Łódzkiej, Łódź 2023, s. 167-172, ISBN 978-83-66741-92-8, doi: 10.34658/9788366741928.26.