NUI Galway – UL Alliance First Annual ENGINEERING AND - ARAN ...

dcat: Searching Public Sector Information Across Borders
Abstract
Opening public sector information has recently become
a trend in many countries around the world. Online
government data catalogues act as one-stop data
portals providing description of available government
datasets. However, these catalogues remain isolated
from each other. Potential benefits from federating
geographically overlapping or thematically
complementary catalogues are not realized. We
propose dcat - an RDF Schema vocabulary - as an
interchange format among data catalogues and as a
way of bringing them into the Web of Linked Data,
where they can enjoy interoperability among themselves
and with other deployed datasets.
1. Motivation
“Open Data” and “Open Government”—these
terms describe a recent trend towards more openness
and transparency in government, a development that has
recently been embraced by some administrations. This
development promises social benefits through increased
transparency and openness, and economic benefits
through realising the full potential of data that has
already been produced as part of the administration's
day-to-day operations and paid for by the taxpayer.
Sharing this data leads to cost savings for the private
sector and enables the provision of new innovative
services that the government cannot or will not provide.
Data catalogues such as data.gov in the US and
data.gov.uk in the UK have recently appeared as onestop
web portals that facilitate access and increase
findability of such data by providing lists of
government datasets along with metadata.
It is common for the data catalogues themselves to
be available in some format that is amenable to machine
processing. We propose a standardised interchange
format for such machine-readable representations of
government data catalogues. The adoption of such a
format enables embedding machine-readable metadata
in web pages which increases findability by nextgeneration
search engines. It also empowers
decentralised publishing and aggregation into national
or super-national (e.g., EU-wide) catalogues. Federated
search over catalogues with overlapping scope, such as
the catalogues for San Francisco, California, and the
entire US becomes possible when interoperable
description of catalogues are available.
2. Model
Based on our survey, we have developed an RDF
Schema vocabulary that allows the expression of data
catalogues in the RDF data model. We have chosen
Fadi Maali, Richard Cyganiak, Vassilios Peristeras
Digital Enterprise Research Institute, NUI Galway
(firstname.lastname)@deri.org
85
RDF because (i) most of the use cases considered in
Section 1 involve querying of aggregated data, which is
well-supported in RDF; (ii) re-use and extension of
existing metadata standards such as Dublin Core is
straightforward in RDF; and (iii) for compatibility with
Linked Data. The figure below shows the basic model.
Full documentation is available online 1 .
3. Adoption
Work on dcat is pursued under the W3C e-gov
Interest Group with a number of stakeholders involved
(http://www.w3.org/egov/wiki/Data_Catalog_Vocabula
ry).
dcat was recently utilized by the Open Knowledge
Foundation in the UK to enable searching public sector
information across Europe through a faceted searching
interface available at: http://publicdata.eu
data.gov.uk also uses dcat in their RDFa
annotation.
Acknowledment
The work presented in this paper has been funded in
part by Science Foundation Ireland under Grant No.
SFI/08/CE/I1380 (Lion-2).
1 http://vocab.deri.ie/dcat-overview