Applying Transformers to Naval Route Prediction

This work studies the application of the transformer technology, usually used for natural language processing tasks, in the realm of maritime route prediction. We use as dataset the reports from the Automatic Identification System which contain the recent historic coordinate of most ships. Then we implement a methodology to discretize the dataset by splitting the world into a grid and splitting the data into routes, so that we can train a model to predict the next element of the route. And finally we perform the training on a subset of the full dataset, where we achieve a 68 % accuracy in the top 25% of ships in the validation dataset.

• Autores: Gómez Zapata, Camilo
• Tipo de recurso: Trabajo de grado de pregrado
• Fecha de publicación: 2024
• Institución: Universidad de Antioquia
• Repositorio: Repositorio UdeA
• Idioma: eng
• OAI Identifier: oai:bibliotecadigital.udea.edu.co:10495/45988
• Acceso en línea: https://hdl.handle.net/10495/45988
• Palabra clave: Deep learning (Machine learning); Aprendizaje profundo (Aprendizaje automático); Spatial data mining; Minería de datos espaciales; Ships - Automatic identification systems; Barcos - Sistemas de identificación automática; Maritime transport; Transporte marítimo; Machine learning; Aprendizaje automático; Artificial intelligence; Inteligencia artificial; Neural networks; Redes neuronales; Naval route prediction; Predicción de Rutas Navales; http://aims.fao.org/aos/agrovoc/c_49834; http://aims.fao.org/aos/agrovoc/c_27064; http://aims.fao.org/aos/agrovoc/c_37467; http://id.loc.gov/authorities/subjects/sh90001937; http://id.loc.gov/authorities/subjects/sh2021006947; http://id.loc.gov/authorities/subjects/sh2019001855; http://id.loc.gov/authorities/subjects/sh2003001275; http://vocabularies.unesco.org/thesaurus/concept11744; ODS 9: Industria, innovación e infraestructura. Construir infraestructuras resilientes, promover la industrialización inclusiva y sostenible y fomentar la innovación
• Rights: openAccess
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/