Centre de documentation
scientifique

A hierarchical model for semantic trajectories and event extraction in indoor and outdoor spaces / Hassan Noureddine (2022)  

Public Titre : A hierarchical model for semantic trajectories and event extraction in indoor and outdoor spaces Type de document : Thèse/HDR Auteurs : Hassan Noureddine, Auteur ; Christophe Claramunt, Directeur de thèse ; Cyril Ray, Directeur de thèse Editeur : Brest : Université de Bretagne Occidentale Année de publication : 2022 Importance : 113 p. Format : 21 x 30 cm Note générale : bibliographie Thèse de Doctorat de l'Université de Bretagne Occidentale, spécialité Géomatique Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique [Termes IGN] approche hiérarchique [Termes IGN] détection d'événement [Termes IGN] données spatiotemporelles [Termes IGN] espace intérieur [Termes IGN] espace urbain [Termes IGN] mobilité humaine [Termes IGN] modèle sémantique de données [Termes IGN] modélisation 3D [Termes IGN] ville intelligente Index. décimale : THESE Thèses et HDR Résumé : (auteur) The interest in exploiting crowd-sourced information has recently emerged as it can bring many valuable benefits for many application domains. This is particularly the case for realtime mobile crowd-sourcing data often available in indoor and outdoor environments. Such data offers many opportunities for analysing human mobility, especially when associated with multidimensional contextual information. Considering the diversity of multi-environment spaces and where mobility occurs, raises several data modelling, management and processing research challenges.When associated with multiple contextual information, indoor and outdoor mobility analysis stresses the need for appropriate and flexible data abstractions at the modelling level to represent the spatial, temporal and semantic data that arise in a smart city environment. While recent approaches often considered this issue using the common stops and moves model, this does not completely cover the multi-dimensional contextual information that arises in real-time on humans navigating through indoor and outdoor spaces. It also increases the need for computing systems and data architectures to process spatio-temporal data in a timely manner when searching for complex mobility events of interest. Despite the ability to represent spatio-temporal events, such systems require well-defined and flexible data manipulation languages to support abstraction and composition mechanisms for analysing urban mobilities.This thesis aims to provide the necessary constructs for analysing mobile crowd-sensed information that arises in indoor and outdoor spaces. In order to better understand urban mobility data in batch and real-time, we consider a broad range of contextual information that can be associated with mobility data. We introduce an indoor and outdoor spatial data model represented as a multi-layered graph and constructed with crowd-sourced trajectory data. The novelty of the approach lies in the fact that it provides a homogeneous and flexible hierarchical indoor and outdoor spatial model that can be associated with crowd-sensed trajectory data on the fly. Our modelling approach defines generic and flexible semantic trajectories considering multiple collaborative data semantics at different granularities and where trajectory segmentation relies on evolving semantic values. This thesis develops a modelling framework for complex events applied to our indoor and outdoor semantic trajectory model based on a formallanguage that establishes the required operations for the composition of the events. We have implemented data pipelines to examine our approach’s efficiency. The whole approach is experimented and applied to participatory data from a real case study to show its suitability, scalability and performance. Note de contenu : 1- Introduction 2- Related work 3- Semantic trajectory data model For indoor and outdoor spaces 4- Composite event extraction from stream of semantic trajectories 5- Conclusion Numéro de notice : 24080 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Thèse française Note de thèse : Thèse de Doctorat : Géomatique : Brest : 2022 Organisme de stage : Institut de Recherche de l’Ecole Navale DOI : sans En ligne : https://theses.hal.science/tel-03888591 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=102293