• AIS
• AIS
• Knowledge discovery
• M-Atlas
• maritime awareness
• mobility data mining
• trajectory analysis
• trajectory clustering
• trajectory generation

This work investigates the use of historical Automatic Identification System (AIS) data as collected from the Maritime Safety and Security Information System (MSSIS) network together with prototype trajectory analysis software developed at the Centro Nazionale di Ricerca (CNR) and the University of Pisa Knowledge Discovery in Databases Lab.
The MSSIS network is a freely-shared, unclassified, near real-time data collection and distribution network for AIS data based on the
contributions of a global network of member nations. The MSSIS is only one example of AIS data sharing networks, and this data was used as a starting point to understand the potential of AIS data for the purpose of trajectory analysis.

This work details some limitations of the MSSIS network for trajectory analysis, but shows several
examples of the utility of data mining techniques on historical maritime data. Data coverage can be
an issue with land-based AIS networks and so this analysis focuses on the Strait of Gibraltar, an area of good MSSIS spatial coverage, in order to focus on the potential of the techniques. The approach taken uses some simple fixes to better separate the data into distinct trajectories, such as an algorithm to detect locations of vessel stops. This allowed for the creation of trajectories which better represent “trips”
from one port or anchorage area to another, having few artificial stops to due gaps in historical data coverage in open water.