Vessel track information mining using AIS data
Feng Deng, Sitong Guo, Yong Sheng Deng, Hanyue Chu, Qingmeng Zhu, Fuchun Sun · 2014
In recent years, vessel traffic and maritime situation awareness become more and more important for countries across the world. AIS data contains much information about vessel motion and reflects traffic characteristics. In this paper, data mining is introduced to discover motion patterns of vessel movements. Firstly, we do statistical analysis for large scale of AIS data. Secondly, we use association rules to analyze the frequent moving status of vessels. We extend the dimensions of data features, improve the algorithm in efficiency and import the concept of time scale in the algorithm based on the previous relative work. Thirdly, we introduce Markov model to make supplement for the association rules. The prediction results in the Markov process are further used to do the anomaly detection. The method in this paper provides novel idea for the research in AIS data and the management of maritime traffic.