=Paper=
{{Paper
|id=Vol-1963/paper461
|storemode=property
|title=SW at the department ‘Omgeving’ of the Flemish Gov
|pdfUrl=https://ceur-ws.org/Vol-1963/paper461.pdf
|volume=Vol-1963
|authors=Paul Hermans
|dblpUrl=https://dblp.org/rec/conf/semweb/Hermans17
}}
==SW at the department ‘Omgeving’ of the Flemish Gov==
https://ceur-ws.org/Vol-1963/paper461.pdf
SW at the department ‘Environment’ of the Flemish
Government
Paul Hermans1
1
ProXML, 3140 Keerbergen, Belgium
Abstract. This project proves that the semantic technology stack and related tool-
ing allow to do data integration in a fast and agile way. Several technologies have
been utilised: Ontology-Based Database Access, federated SPARQL, federation
middleware. The solution is using the RDF Data Cube vocabulary for capturing
the emission observations done. The additional 5 star LOD publishing was easily
achieved at a minimal cost.
Keywords: Data integration, OBDA, federated search, LO(S)D.
1 Background
The ‘Omgeving’ Department is the environmental administration of the government of
Flanders. It is responsible for preparing, following up and evaluating the Flemish envi-
ronmental policy. In Belgium, companies that want to emit polluting substances in the
air or water must have an environmental permit. Some of them are also obliged to report
annually about the emissions of the previous year. In fact, data regarding the amount of
substances per location have been collected since 2004.
The idea was to integrate these data with all kinds of other data sources (internal and
external) to offer:
• the general public an application showing the emitted substances over the years in
the area that they live
• companies the ability to benchmark themselves against other, similar, organisations
• public servants analytics dashboards to gain insights to steer the policy making.
The ultimate aim was that by integrating the data (research) questions can be addressed
not being able to be answered on the separate data silos.
The solution preferably needed to be based on open source software or using open
source libraries.
The project serves also as a pilot site of the OpenGovIntelligence project[1], receiv-
ing funding from the European Union’s Horizon 2020 research and innovation pro-
gramme under grant agreement No 693849, which aims to modernize Public Admin-
istration by connecting it to Civil Society through the innovative application of Linked
Open Statistical Data (LOSD).
2
2 Solution
2.1 Solution outline
Dataintegration. The actual status is that more than 10 data silos have been integrated.
These datasets contain data managed by the department itself (archival systems,
RDBMS) and datasets published by other departments (Flemish addresses database,
Belgian company register) together with well-known classification systems such as
NACE (economical activities) and NIS (administrative geographical entities).
The solution takes a hybrid approach. For some datasets, the data are transformed
via dedicated ETL processes into triples and these are loaded in a triple store. Other
datasets, mainly from existing relational databases, are virtualized via OBDA[2] (On-
tology-Based Data Access). The own ontology and controlled vocabularies will be
managed with VocBench V3[3]. There are also connections to external SPARQL end-
points. In front of all these different SPARQL endpoints we offer a federation layer.
Data model. The most important datasets are in fact observations. Hence our use of the
RDF Data Cube vocabulary[4]: a vocabulary explicitly made to capture statistical data
to allow OLAP operations such as slice and dice, roll-up and drill-down. The code lists
are encoded in XKOS[5], an extension of SKOS for statistical classifications.
LOD Publishing. All entities have their subject pages published using dereference-
able URL’s. Next to this there are also public SPARQL endpoints available. But the
important point is that the LOD publishing was not the first aim of the project: it was a
nice free add-on.
General Public Dedicated Application. An application has been build showing the
emitted substances over the years in the area that they live. This is a traditional web
application using Polymer [6] web components, which is the standard in the depart-
ment. This application is talking to the integrated set of data via the available SPARQL
endpoints.
Business Intelligence. A connector has been developed for exploratory [7], a Data Sci-
ence environment based on RStats [8], to connect with triple stores, so that BI reporting
can be done using the integrated datasets.
2.2 Business Benefits
The approach used has clearly proven that semantic technologies allow to do data inte-
gration in a fast and agile way. Answers can now be given to questions which involve
data from multiple datasets. The fact that 5 star LOD publishing is only one step away
is a nice add-on. The system is used internally for the moment and will be become open
to the public early November.
3
References
1. OpenGovIntelligence, http://www.opengovintelligence.eu/
2. Diego Calvanese, Giuseppe De Giacomo, Domenico Lembo, Maurizio Lenzerini, Antonella
Poggi, Riccardo Rosati, Ontology-based database access, http://www.dis.uni-
roma1.it/~degiacom/papers/2007/sebd07.pdf
3. VocBench V3, http://vocbench.uniroma2.it/
4. RDF Data Cube Vocabulary, https://www.w3.org/TR/vocab-data-cube/
5. XKOS, Extended Knowledge Organization System, http://www.ddialliance.org/Specifica-
tion/RDF/XKOS
6. Polymer Project, https://www.polymer-project.org/
7. Exploratory, https://www.exploratory.io/
8. RStats, https://www.r-project.org/