Web Semantics: Science, Services and Agents on the World Wide Web 41 (2016) 1–8
Contents lists available at ScienceDirect
Web Semantics: Science, Services and Agents
on the World Wide Web
journal homepage: www.elsevier.com/locate/websem
Exploratory spatio-temporal analysis of linked statistical data
Vuk Mijović
a
, Valentina Janev
b,*
, Dejan Paunović
b
, Sanja Vraneš
b
a
School of Electrical Engineering, University of Belgrade, Institute Mihajlo Pupin, Belgrade, Serbia
b
University of Belgrade, Institute Mihajlo Pupin, Belgrade, Serbia
article info
Article history:
Received 11 April 2016
Received in revised form
11 August 2016
Accepted 24 October 2016
Available online 1 November 2016
Keywords:
Linked data
Statistics
Spatio-temporal
Exploration
Visualization
Interoperability
abstract
Publishing and sharing open government data in Linked Data format provides many opportunities in
terms of data aggregation/integration and creation of information mashups. Statistical data, that contains
various performance indicators and their evolution through time, is an example of data that can be
used as the foundation for policy prediction, planning and adjustments, and can be re-used in different
applications. However, due to Linked Data being relatively a new field, currently there is a lack of tools
that enable efficient exploration and analysis of linked geospatial statistical datasets. Therefore, ESTA-LD
(Exploratory Spatio-Temporal Analysis) tool was developed to address some of the Linked statistical Data
management issues, such as crossing the statistical and the geographical dimensions, producing statistical
maps, visualizing different measures, and comparing statistical indicators of different regions through
time. This paper discusses the modeling approach that was adopted so that the published data conform
to the established standards for representing statistical, spatial and temporal data in Linked Data format.
The main contribution is related to the delivery of state-of-the-art open-source tools for retrieving, quality
assessment, exploration and analysis of statistical Linked Data that is made available through a SPARQL
endpoint.
© 2016 Elsevier B.V. All rights reserved.
1. Introduction
Statistical data is often used as the foundation for policy
prediction, planning and adjustments, and therefore has a
significant impact on the society (from citizens to businesses to
governments). In the last few years, with the rise of the open data
movement, a large and increasing number of governments and
organizations have started to make information freely available
and easily accessible online. In order to increase transparency, the
information is also published as Linked Open Data [1].
From the government systems perspective, the Linked Data
approach can be observed as a technique for making the data
interoperable and ready for consumption. In order to harmonize
approaches used for describing the datasets, semantic services or
repositories, the European Commission, in collaboration with the
W3C consortium, has accepted a set of standard vocabularies that
should be used to build public administration services [2]. In the
ISA programme framework, the European Commission supports
*
Corresponding author.
E-mail addresses: Vuk.Mijovic@pupin.rs (V. Mijović), Valentina.Janev@pupin.rs
(V. Janev), Dejan.Paunovic@pupin.rs (D. Paunović), Sanja.Vranes@pupin.rs
(S. Vraneš).
the development of tools, services and frameworks in the area of e-
Government through more than 40 actions.
1
Currently, the JOINUP
repository is used for storing the descriptions of schemata used
in the publicly available datasets, as well as services that enable
access/retrieval of data.
In general, the wider adoption of standards for representing and
querying semantic information, such as RDF(s) and SPARQL, along
with increased functionalities and improved robustness of mod-
ern RDF stores, have established Linked Data and Semantic Web
technologies in the areas of data and knowledge management.
However, these technologies are still quite novel, and a lot of the
tooling and standards are either missing, still in development, or
not yet widely accepted. For example, the GeoSPARQL [3] standard
that supports representing and querying geospatial data on the Se-
mantic Web was published in June 2012, but the Spatial Data on
the Web Working Group is still working on clarifying and formal-
izing the relevant standards landscape with respect to integrating
spatial information with other data on the Web, discovering of dif-
ferent facts related to places, and identifying and assessing existing
methods and tools in order to create a set of best practices. The RDF
1
http://ec.europa.eu/isa/ready-to-use-solutions/index_en.htm.
http://dx.doi.org/10.1016/j.websem.2016.10.002
1570-8268/© 2016 Elsevier B.V. All rights reserved.