Revealing an inherently limiting factor in human mobility prediction - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Emerging Topics in Computing Année : 2022

Revealing an inherently limiting factor in human mobility prediction

Résumé

Predicting how humans move within space and time is a central topic in many scientific domains such as epidemic propagation, urban planning, and ride-sharing. However, current studies neglect individuals' preferences to explore and discover new areas. Yet, neglecting novelty-seeking activities at first glance appears to be inconsequential on the ability to understand and predict individuals' trajectories. We claim and show the opposite in this work: exploration-like visits strongly impact mobility understanding and anticipation. We start by proposing a new approach to identifying exploration visits. Based on that, we construct individuals' mobility profiles using their exploration inclinations-Scouters (i.e., extreme explorers), Routiners (i.e., extreme returners), and Regulars (i.e., with no extreme behavior). Finally, we evaluate the impacts of novelty-seeking, quality of the data, and the prediction task formulation on the theoretical and practical predictability extents. The results show the validity of our profiling and highlight the obstructive impacts of noveltyseeking activities on the predictability of human trajectories. In particular, in the next-place prediction task, from 40% to 90% of predicted locations are wrong, notably with Scouters.
Fichier principal
Vignette du fichier
paperINRIAJournal_IEEE_FINAL_version.pdf (6.73 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03905517 , version 1 (18-12-2022)

Licence

Paternité

Identifiants

Citer

Licia Amichi, Aline Carneiro Viana, Mark Crovella, Antonio A. F Loureiro. Revealing an inherently limiting factor in human mobility prediction. IEEE Transactions on Emerging Topics in Computing, In press, ⟨10.1109/TETC.2022.3229088⟩. ⟨hal-03905517⟩
39 Consultations
50 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More