Predicting Next Whereabouts Using Deep Learning

Ana Paula Galarreta, Hugo Alatrista-Salas, Miguel Nunez-del-Prado

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

Resumen

Trajectory prediction is a key task in the study of human mobility. This task can be done by considering a sequence of GPS locations and using different mechanisms to predict the following point that will be visited. The trajectory prediction is usually performed using methods like Markov Chains or architectures that rely on Recurrent Neural Networks (RNN). However, the use of Transformers neural networks has lately been adopted for sequential prediction tasks because of the increased efficiency achieved in training. In this paper, we propose AP-Traj (Attention and Possible directions for TRAJectory), which predicts a user’s next location based on the self-attention mechanism of the transformers encoding and a directed graph representing the road segments of the area visited. Our method achieves results comparable to the state-of-the-art model for this task but is up to 10 times faster.

Idioma originalInglés
Título de la publicación alojadaModeling Decisions for Artificial Intelligence - 20th International Conference, MDAI 2023, Proceedings
EditoresVicenç Torra, Yasuo Narukawa
EditorialSpringer Science and Business Media Deutschland GmbH
Páginas214-225
Número de páginas12
ISBN (versión impresa)9783031334979
DOI
EstadoPublicada - 2023
Evento20th International Conference on Modeling Decisions for Artificial Intelligence, MDAI 2023 - Umeå, Suecia
Duración: 19 jun. 202322 jun. 2023

Serie de la publicación

NombreLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volumen13890 LNCS
ISSN (versión impresa)0302-9743
ISSN (versión digital)1611-3349

Conferencia

Conferencia20th International Conference on Modeling Decisions for Artificial Intelligence, MDAI 2023
País/TerritorioSuecia
CiudadUmeå
Período19/06/2322/06/23

Huella

Profundice en los temas de investigación de 'Predicting Next Whereabouts Using Deep Learning'. En conjunto forman una huella única.

Citar esto