Heterogeneous data integration and spatial reasoning for locating victims in mountain areas – CHOUCAS
Heterogeneous data integration and spatial reasoning for localizing victims in mountain area
The project CHOUCAS is interdisciplinary involving researchers in geographic information sciences, computer science and automatic language processing. The aim of the project is to provide methods and tools for enriching, structuring, querying and visualising spatial data coming from heterogeneous sources and reasoning to help the rescue teams to speed up the phase of victims location and to better estimate a probable location area.
Proposal of methods and tools for locating victims in mountain areas
1 / Formalizing the reasoning of rescue process and data needed for the process of localization of victims in the mountains and propose a method suggesting the next question to ask the victim to improve the collection of location indices and thus facilitate decision-making process.<br />2 / Define methods for structuring data from heterogeneous textual sources in order to enrich existing topographic data and produce data for mountain areas.<br />3 / Define methods to transform a relative location in an absolute location and propose a fuzzy model to merge the absolute locations thus obtained to identify a likely presence area.<br />4 / Define geovisualization methods for imperfect data allowing the use of geovisual analysis as a support for spatial reasoning in a decision-making context.<br />5 / Design a generic service-oriented mediation platform model to allow the integration of heterogeneous data sources, namely the data produced by the CHOUCAS project and the data available on the Web that may be conflicting.
1 / Methods for the identification of different named spatial entities
2 / Method of disambiguation of named spatial entities
3 / Method of matching named spatial entities with GPS traces
4 / Proposal of a model for representing relative locations
5 / Method of spatialization of the location element «I am under a power line«
1 /A first prototype for decision making for locating victims in mountain
2 / Specification of the alpha version of the DSL representation of data adapted to the annotation of named spatial entities
3 / Specification of the first version of a DSL that proposes a taxonomy of services classified according to their functionality
4 / A Service Integration Demonstrator for the Recognition and Annotation of Named Spatial Entities
5 / -QGIS open source plug-in allowing to facilitate the access and the visualization of data of point type (shelters, bodies of water ...) and of type line (trek routes), coming mainly from collaborative sites ( github.com/ANRChoucas)
Continue the work with emphasis on evaluation and genericity.
Ana-Maria Olteanu-Raimond, Paule-Annick Davoine, Mauro Gaio, Eric Gouardères, Marie-Dominique Van Damme et al., Projet CHOUCAS : Intégration de données hétérogènes et raisonnement spatial pour l'aide à la localisation des victimes en montagne. Spatial Analysis and GEOmatics 2017, Rouen, France, 3-6 Novembre 2017. hal-01649156
Medad A., Gaio M., Mustière S. « Appariement automatique de données hétérogènes: textes, traces GPS et ressources géographiques. » Papier accepté Sagéo'2018, Montpellier, 6-9 Novembre 2018.
Gaio M., Moncla L. « Geoparsing and geocoding places in a dynamic space context: The case of hiking descriptions. »(à paraître), chapitre de livre dans ‘’The Semantics of Dynamic Space in French’’, M. Aurnague & D. Stosic (Eds.), pp.487-530, collection «Human Cognitive Processing« (HCP), Cognitive Foundations of Language Structure and Use J. Benjamins.
Moncla, L. Gaio, M. (accepté) Services Web pour l'annotation sémantique d'information spatiale à partir de corpus textuels. (22 pages) dans Revue Internationale de Géomatique, Lavoisier.
Mattia Bunel, Ana-Maria Olteanu-Raimond, Cécile Duchêne, Référencement spatial indirect : modélisation à base de relations et d’objets spatiaux vagues. Papier accepté Sagéo'2018, Montpellier, 6-9 Novembre 2018.
Halilali M., Gouardères E., Devin F., Gaio M. «Plateforme logicielle pour l’intégration et la composition de services géospatiaux » Papier accepté Sagéo'2018, Montpellier, 6-9 Novembre 2018.
Mattia Bunel, Cécile Duchêne, Ana-Maria Olteanu-Raimond, Modélisation et raisonnement spatial flous pour l'aide à la localisation de victimes en montagne. 27es journées de la Recherche , Cité Descartes, Marne-la-Vallée, les 22 et 23 mars 2018. <hal-01806891>
Amine Medad, Mauro Gaio, Sébastien Mustière, Localisation automatique des points d'intérêt par la mise en correspondance du texte narratif et de la trace GPS. 27es journées de la Recherche , Cité Descartes, Marne-la-Vallée, les 22 et 23 mars 2018.
More than 15,000 mountain rescue requests are recorded each year on the French territory. Despite the technological advances of positioning systems and the growing number of geo-applications, the location of the victims during the alert phase remains crucial to the rescue public services teams. Data heterogeneity, victim’s mountain poor knowledge and the imprecision of the travelled itinerary described by the victims are some factors that make this step difficult. The project CHOUCAS is an interdisciplinary project involving researchers from different domains such as data and knowledge management, retrieval information, spatial data reasoning, and spatial data visualization. The aim of the project, in response to the needs expressed by the Peloton de Gendarmerie de Haute-Montagne de Grenoble, is to propose: methods and tools to build and enhance spatial data coming from heterogeneous sources, fuzzy spatial reasoning models and methods, and geo-visualization tools to improve the decision making process regarding victims’ localisation in mountain areas.
Project coordination
Ana-Maria RAIMOND (Institut National de l'Information Géographique Forestière)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Partner
LIG Laboratoire d'informatique de Grenoble
IGN-COGIT Institut National de l'Information Géographique Forestière
LIUPPA Université de Pau et des Pays de l'Adour
Help of the ANR 591,190 euros
Beginning and duration of the scientific project:
March 2017
- 42 Months
