Classifying Vessels Operating in the South China Sea by Origin with the Automatic Identification System
Abstract
This research focuses on building classification models with multinomial responses based upon seven months of Automatic Identification System (AIS) data gathered from the South China Sea. The models, built using Gradient Boosted Machines (GBM), assess the validity of utilizing AIS to confirm an operating vessels origin, by country and geographical region. Two types of models are built. The first model captures the naturally dependent nature of AIS signals and serves as a proof of concept for how well a global model trained over many years could perform The second model attempts to reduce the dependency between AIS signals in order to characterize maritime patterns of behavior by country and region. With relative accuracy, both types of models are able to predict a vessels origin and provide insight into maritime patterns of behavior.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2018
- Accession Number
- AD1052543
Entities
People
- Kimberly M. Cull
Organizations
- Naval Postgraduate School