Introduction
The “OSLO pilot” is set up by the OSLO consortium in collaboration with the Flemish eGovernment Coordination Unit (CORVE) and the Flemish Geographical Information Agency (AGIV).
The portal is publishing the Belgian Business register (kruispuntbank voor ondernemingen) and inter-links data from the Regional address register CRAB (centraal referentie adressenbestand), containing a geographical positioning for each of the 3 million+ addresses in Flanders.
This pilot was carried out in the context of OSLO 1.1 and demonstrates the value of ‘Linked Data’ to inter-linke and publish disparate crossroad databases.
This pilot is an initiative of V-ICT-OR (The Flemish Organization for ICT in Local Government, Belgium) and was realized in collaboration with the iMinds Research Groups MMLab and MICT, Ghent University
Target audience
The potential targets are all different types of stakeholders that partake in the public service provision process, and would like to explore the benefits of the OSLO Open Standards.
Targeted benefits
This case study explains how organisations and their addresses can be described in RDF using the OSLO Vocabulary.
This approach aims to create (1) a better harmonization of the authentic data and (2) a better adoption of the authentic data, through the use of an open and extensible data format.
Context
The OSLO Vocabulary is a simplified, reusable and extensible data model that captures the fundamental characteristics of information exchanged by public administration in the domains: contact information, localisation and public services.
The standards of the Flemish OSLO project are local extensions of the core Person, Business, Location, and Public Service vocabularies created at European level in the context of the ISA Programme (Interoperability Solutions for European Public Administrations) of the European Union. These four core vocabularies are simplified, reusable, and extensible specifications for information exchange.
This pilot demonstrates how organizations and addresses can be described in RDF, using the (1) OSLO Business and (2) OSLO Location Vocabularies.
Both Vocabularies abide by the Linked Data Principles
- they promote the use of common identifiers for organizations and addresses in the form of URI’s.
- they can be easily combined with other Linked Data vocabularies
- they can easily be extended with new classes and attributes to fulfil new domain specification.
Businesses
This pilot is using data from the authentic register that contains all businesses in northern Belgium area (Flanders): VKBO (verrijkte kruispuntbank voor ondernemingen).
VKBO is able to exchange data from businesses, through inter-linking them to a unique identifier. This enables the integration of several disparate datasets: the data from the Federal company register (Kruispuntbank van Ondernemingen - Banque-Carrefour des Entreprises) and other crossroad databases including the crossroad database of the National Social Security (RSZ), consultation of annual accounts ( the annual accounts or consolidated annual accounts filed with the National Bank since 1992) and permits.
The Business Vocabulary, rebranded into the Registered Organisation Vocabulary[1], enables anyone to describe the essential elements of a registered organisation.
- the legal name of the organization
- the registered number of the organization
- the legal address of the organization
- the activities for which the organization is registered for
- the type of organization
Each organization is identified by a unique URI, which is alligned with the URI strategy from the Flemish Government and the European directive 2012/17/EU.
The ORG Ontology is more generic and makes it possible to describe organizational structures using RDF:
- the organizational decomposition of an organization
- the people who are members of the organization at each one of its levels
Addresses
This pilot is using data from the regional register that contains all valid addresses in northern Belgium area (Flanders): CRAB. Nowadays, all addresses are decentrally updated and maintained by the local cities. The CRAB database is built according to international Inspire-rules. CRAB stands for ‘Centraal Referentie Adressen Bestand’ (= Central Reference Address Database). In addition to house numbers and street names, the database also contains a geographical positioning for each of the 3 million+addresses in Flanders. The CRAB is an initiative, primarily concerned with exchange and quality of address details.
The Location Vocabulary enables anyone to describe the essential elements of an address.
The INSPIRE Data Specifications are legally binding. For addresses, Core Location can be seen as a subset of the INSPIRE address specification.
The overall concept of this data specification is that an address has a “locator” of a particular type (e.g. an address number that enables a user to distinguish it from the neighbouring addresses); and a geographic position, which enables an application to locate the address spatially. To identify the address unambiguously in a wider context an address must be associated with a number of “address components” that define its location within a certain geographic area. Each of the address components represents a spatial identifier (for example the name of a road, municipality).
‘Linking’ of both datasets
The source of the address data related to the businesses is the Belgian Federal Government: the Belgian company register (Kruispuntbank van Ondernemingen - Banque-Carrefour d’ Entreprises).
Address information of the Federal Government and the regional Governments (Flemish Government) are housed in isolated registries, thy have an heterogeneous data format and there is a lack of common identifiers.
In order to inter-link the authentic data from the Business Register and the authentic source of the addresses (CRAB), AGIV maintains a ‘mapping’ between the federal addresses and the unique regional addresses (CRAB). This ‘mapping’ table is published by AGIV as a webservice: CRAB-WST.
Getting started
To browse the data served by the pilot, you may visit http://purl.org/oslo/pilots/business_addresses (expected 27/5/2013)
There are three ways to reuse the data programmatically:
1) query access via SPARQL (using Linked Data Fragments),
2) entry level access (via HTTP),
3) via datadumps.
All data is contained in the RDF model where all facts are described as “triples”. More information about RDF and ways to handle this data can be found at: http://linkeddatabook.com/editions/1.0/#htoc75 and http://euclid-project.eu/
Query access via SPARQL (1) is available via client-side libraries thanks to Linked Data Fragments.
Entity level access is possible using the URIs included in the datadump.
To get access to the data, you may visit http://purl.org/oslo/pilots/business_addresses/data/ (expected 27/5/2013)
Terms and conditions
The OSLO pilot, set up by the OSLO consortium in view of the development of the OSLO standards.
There is no strict guarantee that the pilot is available. The service can be interrupted temporarily for maintenance operations of any kind. However, the OSLO consortium commits itself to keeping the system available.
Because we are dealing with a test environment, there is no guarantee of the services/applications being available, or the quality of the data.
The OSLO consortium is not responsible and cannot be held liable for possible problems, conflicts, direct or indirect damage resulting from the use of the pilot.
OSLO Vocabulary Working Group
The specification is developed by a multi disciplinary Working Group, with a total of 58 people from 28 organisations and representatives of the ISA programme.
The OSLO specification is the result of a public-private partnership initiated by V-ICT-OR, the Flemish Organization for ICT in Local Government, and funded by Flemish ICT service providers (e.g. BCT, CEVI,, Remmicom and Schaubroeck) and public administrations (e.g. CORVE, Digipolis).
The research activities fort this pilot were funded by the OSLO consortium and the Flemish department of Economics, Science and Innovation (EWI - Open Data project). A special thanks to Paul Hermans for his support and Kelly Bonneure fo the preparation of their datasets.
To express your interest in participating in the public-private partnership email: raf [dot] buyle [at] v-ict-or [dot] be.
References
Open Standards for Linked Administrations in Flanders 1.0
OSLO 1.1 RDF ontology (release candidate)
http://purl.org/oslo/ns/localgov
Case Study: How to describe Organizations in RDF using the Core Business Vocabulary
and the Organization Ontology?
LinkedData@vlaanderen: uri strategie en
vocabularyrichtlijnen - versie 1.0
http://www.opendataforum.info/files/URI_strategie.pdf
Core Location Pilot: Interconnecting Belgian National and Regional Address Dara
Linked Data Fragments (LDF)
http://linkeddatafragments.org
[1] Since 8 January 2013, the Core Business Vocabulary has been formally published on the W3C standards track. It has been revised and renamed into Registered Organization Vocabulary (RegOrg).