Back to top

Link Discovery for Maritime Monitoring


Link discovery in the maritime domain is the process of identifying relations - usually of spatial or spatio-temporal nature - between entities that originate from different data sources. Essentially, link discovery is a step towards data integration, which enables interlinking data from disparate sources. As a typical example, vessel trajectories need to be enriched with various types of information: weather conditions, events, contextual data. In turn, this provides enriched data descriptions to data analysis operations, which may lead to the identification of hidden or complex patterns, which would otherwise not be discovered, as they rely on data originating from disparate data sources.

This chapter presents the fundamental concepts of link discovery relevant to the maritime domain, focusing on spatial and spatio-temporal data. Due to the processing-intensive nature of the link discovery task over voluminous data, several techniques for efficient processing are presented together with examples on real-world data from the maritime domain.


World Borders Shapefile


The shapefile TM_WORLD_BORDERS-0.3 has been downloaded from thematicmapping.org. It is used in this chapter for visualization purposes.


Natural Earth II with Shaded Relief


This file can be downloaded from NaturalEarthData. It is used in this chapter for visualization purposes.


Heterogeneous Integrated Dataset
for Maritime Intelligence, Surveillance and Reconnaissance


This file can be downloaded from NaturalEarthData. It is used in this chapter for visualization purposes.

A list of 15 contextual and 3 surveillance data sets is available online as shapefiles here (H2020 datAcron community at Zenodo). These data sets have been exploited in the examples and the evaluation of link discovery tasks in this chapter.


Contextual Maritime Dataset (RDF Triples)


The Contextual data sets used in the link discovery tasks of this chapter are accessible online. The set of triples have been generated from the above list of data sets using RDF-Gen w.r.t. the datAcron ontology.

RDF Resource Description Framework Flyer Icon


Link Discovery on AIS and Contextual Datasets


The link discovery tasks presented in this chapter have been used for the detection of links (available online). Specifically, this data set contains the links discovered between AIS synopses (available online) and contextual data sets (available online). Specifically, the detected re lations and the contextual data sets used are:
  1. C1 ports of Brittany, World Port Index and SeaDataNet fishing ports for proximity relation "nearto", stored in AIS_nearto_ports.ttl.7z
  2. C4 fishing areas (European Commission) for "within" relation stored in AIS_within_fishingAreas.ttl.7z
  3. C5 fishing interdiction for "within" relation stored in AIS_within_FishingConstraints.ttl.7z
  4. C5 Natura2000 for "within" relation stored in AIS_within_natura2000.ttl.7z
Please notice that the above data sets contain only the detected links. Description of the resources is provided in the corresponding TTL files.

RDF Resource Description Framework Flyer Icon


RDF Triples of Raw and Trajectory Synopses
over AIS Kinematic Messages in Brest


The surveillance data set used for the evaluation of link discovery tasks in this chapter are available online. For the readers' convenience, we provide both TTL and shapefile files for each data set. These data sets have been generated from the trajectory synopses (available online) using RDF-Gen and the datAcron ontology.

RDF Resource Description Framework Flyer Icon


Total Number of Reported Positions During January 2016


This data set has been used as an example of data skew. The data set is available both as a csv file and as a shapefile.


Example of Blocking Italy as a Single Entity


Example of blocking Italy as a single target entity (extracted from world border data set and available online) in a grid of 0.5x0.5 degree granularity. Red cells indicate blocking using MBR (available online as a csv file), while green cells illustrate blocking by interpolation (available online as a csv file).


Example of Blocking a Trajectory


This data set allows the reconstruction of the example of blocking a trajectory. In the corresponding figure, green cells indicate blocking by geometry, while red cells indicate blocking by bounding box. Granularity of grid is 0.5x0.5.