Ontology-Driven GIS


Mohamed Ben Ellefi, Pierre Drap

In recent years, an increasing number of works have shown interest in the development of semantic web based approaches, technologies and tools for supporting life science applications, opening the door for many interesting perspectives. In this context, our research activities focus on the adoption of semantic web technologies in dealing with data serialization and GIS data-mining. Semantic web vision for data serialization intends mainly to fill the gap between machine and human readability. To achieve this goal, a source data goes through different stages to be published in semantic web format.

We propose an ontology-based method for a semantic data life cycle that starts by extracting and transforming data sources from various formats (image, xml, etc) into the triples format, such as Ontology Web Languages (owl) and Resources Description Framework (RDF). The next step consists of assigning namespaces to all the resources in this triple data in order to make them accessible via their URIs. Then, we develop an ontology that model this transformed data. In the ontology modelling steps, we adopt W3C recommendations by reusing existing terms whenever possible, i.e. searching existing terms in the Linked Open Vocabularies (http://lov.okfn.org/). Once the dataset is modelled and available in semantic web format, we offer a GUI interface where we can query this data with semantic web query languages as SPARQL and SQWRL. Furthermore, our research activities intend to extend the querying GUI tool by adding more built-ins rules in order to allow users to retrieve answers to complex questions. Further step in this life cycle vision consists of linking the transformed dataset to other existing datasets for better data integrity and interoperability.

In the 3D amphorae context, we are working on the development of GUI interfaces that facilitate the data-mining of spatial information for different amphorae in the location wreck [1]. In fact, that consists in the implementation of a new approach that allows archaeologist to perform sophisticated queries in human languages over the transformed datasets in semantic web formats.


Recent publication for our works in progress:
Drap, P., Papini, O., Sourisseau, J. C., & Gambin, T. (2017, May). Ontology-Based Photogrammetric Survey in Underwater Archaeology. In European Semantic Web Conference (pp. 3-6). Springer, Cham.