Web page : Camille BERNARD Web page
Associate Professor at Univ. Grenoble Alpes – Since 2020 September, 1st
Researcher at LIG (STeamer group) & teacher at IAE-Grenoble master MSI.
ATER IUT2 Grenoble – 2019-2020
Research and teaching assistant at LIG (STeamer) and IUT2 Grenoble, Information-Communication and Techniques de Commercialisation departments.
PhD Student – 2015-2019
PhD Defense: November 27, 2019 10am (Maison Jean Kuntzmann)
- Subject: Immersing evolving geographic divisions in the semantic Web: Towards spatiotemporal knowledge graphs to reflect territorial dynamics over time
- Keywords: : Semantic Web ; Spatiotemporal Graph ; Geographic Divisions ; Territorial Statistical Information ; Web of Data ; Spatiotemporal Ontology ; Decision support.
Today, Statistical Institutes (SIs) publish more and more socio-economic statistics on the Web to comply with Open Data Directives so that experts, journalists, and even citizens can use them to conduct various analyses on many countries, or regions. For instance, the evolution of the unemployment rate in a given administrative region can be observed by comparing data available for this region, at two different periods of time.
However, one unexpected underlying problem is that data are often simply not comparable. The rationale for this is that the number of inhabitants, or the boundaries of a region, may have changed over time. This phenomenon is very frequent in Europe (for instance in France, in 2016, administrative regions have been merged into greater regions) or in the U.S.A. (through a well-known process called gerrymandering). Then, data collected in different geographic divisions are not comparable due to potential differences between the observed geographic entities.
Our solution to this problem is to build semantic graphs of Linked statistical Data as well as Linked Data representations for successive versions of the geographic divisions. Each version is connected to its successor through what we call “bridges”. Bridges semantically describe the territorial changes over time and hold the answer to the questions: Where and When the changes occur, Why (because of a reform, because of electoral concerns, etc.) and How divisions change (by fusion of districts, complete redistribution, etc.). Thus, these semantic bridges are meant to help stakeholders understand the evolution of statistics together with the evolution of divisions.
- Funding: Region Rhône Alpes, Allocation Doctorale de Recherche (ARC 7 Innovations, Mobilités, Territoires et Dynamiques Urbaines)
- Start: 2015 October, 1st
Engineer – From November 2012 to September 2015
- Project: ESPON M4D
- Keywords: spatial database, metadata, INSPIRE, long-term series, interoperability, geographic web services
- Summary: Since November 2012, as part of the ESPON M4D project, I have been involved in the development of an INSPIRE compliant data infrastructure (http://database.espon.eu/db2/). This infrastructure is based on the INSPIRE data model extended by the STeamer team to further describe the specificities of territorial statistical data. We develop tools in JAVA, the Web Portal with JavaEE technologies and the database with the PostgreSQL-PostGIS RDBMS. To ensure the infrastructure interoperability, we develop OGC web services prototypes (CSW, WFS, WMS) thanks to open source software such as Geoserver.
- Skills: software development, JAVA, JavaEE, PostgreSQL, PostGIS, Geoserver, Maven