Facial reconstruction

Search LJMU Research Online

Browse Repository | Browse E-Theses

A spatial-temporal data mining method for the extraction of vessel traffic patterns using AIS data

Yang, J, Bian, X, Qi, Y, Wang, X, Yang, Z and Liu, J (2024) A spatial-temporal data mining method for the extraction of vessel traffic patterns using AIS data. Ocean Engineering, 293. ISSN 0029-8018

[img]
Preview
Text
A spatial-temporal data mining method for the extraction of vessel traffic patterns using AIS data.pdf - Published Version
Available under License Creative Commons Attribution.

Download (24MB) | Preview

Abstract

Current traffic pattern mining methods fail to incorporate the temporal co-occurrence of traffic characteristics. To address this problem, a new spatial-temporal data mining method is developed involving three steps. Firstly, a three-dimensional traffic tensor is constructed utilizing AIS data. The AIS data is discretized and numbered so that each AIS data entry is represented by a one-dimensional array that includes region, time, ship type, and speed numbers. Then the AIS array is mapped to the three-dimensional ship traffic tensor. Second, non-negative tensor factorization (NTF) is used to break down the tensor into multiple sub-tensors (i.e., traffic patterns). The effect of the tensor rank (i.e., the number of traffic patterns) is discussed, and the appropriate value of the tensor rank is determined. Thirdly, the traffic patterns are derived from the three-dimensional traffic tensor. The ship traffic pattern is subsequently analyzed in accordance with the actual circumstances. To demonstrate the feasibility of the method, 9 traffic patterns are obtained from the AIS data of Tianjin port-Caofeidian waters. These patterns reveal the presentation of the spatio-temporal distribution of traffic activities of different ship types, and the distribution of navigation speed of different ship types in space, that are of strategic values for port planning, and maritime safety and sustainability.

Item Type: Article
Uncontrolled Keywords: 0405 Oceanography; 0905 Civil Engineering; 0911 Maritime Engineering; Civil Engineering
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
T Technology > TC Hydraulic engineering. Ocean engineering
Divisions: Engineering
Publisher: Elsevier
SWORD Depositor: A Symplectic
Date Deposited: 30 Jul 2024 10:22
Last Modified: 30 Jul 2024 10:30
DOI or ID number: 10.1016/j.oceaneng.2023.116454
URI: https://researchonline.ljmu.ac.uk/id/eprint/23821
View Item View Item