Saturday 01 March 2025
A recent study has shed new light on the effectiveness of using Automatic Identification System (AIS) data for classifying ships into fishing or non-fishing vessels. AIS, a technology used by many commercial and recreational boats to transmit location information, has been gaining attention in recent years as a potential tool for improving maritime domain awareness.
Researchers at the University of Carlos III of Madrid have developed an approach that uses machine learning algorithms to analyze AIS data and identify patterns characteristic of fishing ships. The team’s method involves extracting features from the kinematic data provided by AIS reports, such as speed, course variation, and time spent in a particular area. These features are then used to train a classifier to distinguish between fishing and non-fishing vessels.
The study’s findings suggest that even with minimal information, the proposed approach is able to achieve high accuracy rates for classifying ships into these two categories. This is significant because it highlights the potential of AIS data as a cost-effective and efficient means of monitoring ship activity, which can be particularly useful in combating illegal fishing practices.
One of the key benefits of this approach is its ability to identify vessels that do not transmit AIS reports at all, or only partially report their location information. This is often the case with fishing vessels, which may deliberately avoid transmitting AIS data in order to avoid detection. By analyzing patterns and trends in AIS data, even incomplete reports can be used to infer a vessel’s activity.
The researchers also experimented with different algorithms for balancing class imbalance, where there are significantly more instances of non-fishing vessels than fishing vessels. They found that using techniques such as oversampling and undersampling can improve the performance of the classifier, particularly when dealing with unbalanced datasets.
AIS data has been increasingly recognized as a valuable resource for improving maritime domain awareness. The technology provides location information on ships and other vessels in real-time, which can be used to enhance situational awareness and support decision-making. In addition to classifying ship activity, AIS data can also be used for tracking vessel movement patterns, identifying potential security threats, and monitoring environmental impact.
The study’s findings have implications not only for the maritime industry but also for policymakers and law enforcement agencies seeking to combat illegal fishing practices. By leveraging AIS data and machine learning algorithms, it may be possible to develop more effective and efficient methods for detecting and deterring illegal fishing activities.
Overall, this research demonstrates the potential of AIS data as a powerful tool for improving maritime domain awareness and supporting sustainable fishing practices.
Cite this article: “Classifying Fishing Vessels with Machine Learning and Automatic Identification System Data”, The Science Archive, 2025.
Ais, Machine Learning, Classification, Fishing Vessels, Non-Fishing Vessels, Maritime Domain Awareness, Illegal Fishing Practices, Vessel Tracking, Environmental Impact, Sustainable Fishing Practices
