Demonstration Scenario (Review Year 1)


Instead of having two scenarios, we believe that a single scenario with a convincing narration, plethora of datasets, and multiple queries relating to realistic value added services, would serve our purpose better.

Further, we should also use the same scenario in the following reviews, making it easier to showcase progress. Starting from the state of the art when the project started (baseline), we should be able to present our progress each year and convince the reviewers both for our performed work and for the significance of our contributions:
  • GIS domain: typical geospatial/mapping software and workflow (difficulties/challenges in combining various data sources)
  • Semantic Web domain: GeoSPARQL available, but extremely limited/nonexistent capabilities for geospatial management

The selling point of GeoKnow: streamline the use of geospatial data through Semantic Web technologies, offering readily available services for querying heterogeneous data. In our case, we must convince the reviewers that: (a) the standard GIS-based approach for developing an app would require significant effort, know-how and resources, (b) semantic technologies lower the entry bound for developers and can provide readily available value added services.

Consequently, we should emphasize the diverse nature and origin/use of the datasets (e.g. from census, environmental protection & preservation), lack of interlinking, challenges for Web developers (i.e. not acquainted to GIS tools) to collect/manage/query the data in order to develop a simple web/mobile app.

The demonstration datasets should combine several data sources, so that we are able to demonstrate:
(a) specific working examples of the developed tools, highlighting their usefulness and introducing (where relevant) our future work
(b) diverse, open, multilingual EU data combined together, along with proprietary data
(c) several queries that relate to simple web/mobile applications for SMEs/data economy stakeholders
(d) the significance of applying Semantic Web technologies in the geospatial domain
(e) the differences from current practices and the tangible benefits we introduce in reuse of open/proprietary data for value added services


We consider a tourist who wants to visit (an area of) a country for her summer holidays. She would like to search about hotels that match her needs/taste/financial status, so should would like to find out information about pricing, facilities, distance from port/airport, proximity to town or leisure venues etc. She would also like to find out about museums / cultural events / sport activities during the particular period of her visit, and how close all these would be from the candidate hotels. Moreover, she would like to know where to go for swimming (prefer blue-flag beaches, but also verify quality of water).


  • Data produced by Unister (e.g. hotel sites?, travel data?, statistics?)
  • DBPedia
    • Municipalities, protected areas, historic places, protected areas
  • OSM
    • Beaches, hotels, etc
    • Polygons for several features
  • GeoNames
  • Open German Data?

Demonstration roadmap


In the first step of the demonstration, Sparqlify/LGD and TripleGeo will be demonstrated. Sparqlify can focus on OSM data transformation, while TripleGeo can be exhibited on data, transforming shapefiles and/or the contents of a conventional spatial db (e.g. postgis).


GeoLift will be applied on the most suitable of the above datasets (geonames, dbpedia, OSM, suggestions?) to produce explicit geospatial information within the RDF datasets.
Next, LIMES can be applied to produce interlinked datasets (maybe demonstrating different functionality on different datasets, depending on the exact datatypes upon interlinking will be performed.


FAGI_tr (FAGI for transformation of geospatial RDF) will be applied to align spatial features representations of different datasets (e.g. to transform them into the standard GeoSPARQL format).
FAGI_fu (FAGI for geospatial fusion) will be used to demonstrate some initial geospatial entities fusion functionality (keep average geometry, keep most complex geometry, resize geometry).

Semantic authoring and analysis

Ontowiki will be applied to author enriched/interlinked/fused data.
CubeViz will be applied to produce interesting statistics on the produced RDF data, w.r.t. the demonstration scenario and visualize them (e.g. number of customers books in specific hotels the last years).

Visualization and querying

Facete will be use to visualize the produced data on maps.
Virtuoso will be used to demonstrate efficient querying for interesting querying w.r.t. to the scenario (e.g., find 4-star hotels that are within 1km distance from blue flag beaches


These are suggestions; more queries can be extracted by carefully examining the available datasets

  • Find hotels within 5km from blue flag beaches
  • Find ancient theaters within 10km from 4star hotels
  • Find blue flag beaches with approved water quality that are within a protected area
  • Find tavernas cuisine restaurants and/or villages with a population lower from 1.000 citizens within 2km from my route
  • Calculate population density and number of restaurants around a 10km radius of blue flag beaches