<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>The European Data Portal: Scalable Harvesting and Management of Linked Open Data</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Fabian Kirstein</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff2">2</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Simon Dutkowski</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Benjamin Dittwald</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Manfred Hauswirth</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
          <xref ref-type="aff" rid="aff1">1</xref>
          <xref ref-type="aff" rid="aff2">2</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Fraunhofer FOKUS</institution>
          ,
          <addr-line>Berlin</addr-line>
          ,
          <country country="DE">Germany</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>TU Berlin, Open Distributed Systems</institution>
          ,
          <addr-line>Berlin</addr-line>
          ,
          <country country="DE">Germany</country>
        </aff>
        <aff id="aff2">
          <label>2</label>
          <institution>Weizenbaum Institute for the Networked Society</institution>
          ,
          <addr-line>Berlin</addr-line>
          ,
          <country country="DE">Germany</country>
        </aff>
      </contrib-group>
      <abstract>
        <p>The European Data Portal fosters the adoption and distribution of Linked Open Data by o ering more than 130 million RDF triples. It applies DCAT-AP, the RDF vocabulary for public sector datasets in Europe. Many Open Data solutions do not satisfactorily support this metadata speci cation. To address this problem, we designed and implemented a novel platform for harvesting and managing native DCAT-APcompliant RDF. Our approach uses a triplestore as main database and applies state-of-the-art development and deployment patterns to ensure performance and scalability.</p>
      </abstract>
      <kwd-group>
        <kwd>DCAT-AP</kwd>
        <kwd>Open Data Scalability</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>-</title>
      <p>F. Kirstein et al.</p>
      <p>RDF and SPARQL). If necessary, the acquired metadata is transformed into
DCAT-AP-compliant RDF. The Registry supports the resource-based
management of the major DCAT-AP entities via a RESTful API (Catalogs, Datasets and
Distributions). It receives data from the Harvester, preprocesses it and stores the
RDF in the triplestore. The preprocessing includes generation of consistent URIs,
validation based on SHACL and creation of a full-text search index. The results
of the validation are represented by the Data Quality Vocabulary (DQV)9. The
Quality Service fetches the metadata and validation results, performs availability
checks for external links and creates detailed quality reports. Fig. 1 illustrates
the highlevel architecture, including the most essential services.
The system makes extensive use of an asynchronous and reactive programming
paradigm, applies the container technology Docker10 for deployment and
supports container-orchestration, like Kubernetes11. Therefore, we were able to
decrease the hardware workload and increase the overall performance, in
comparison to the CKAN-based solution. A full harvesting run was accelerated by a
factor 30, where the major blocking factor is the limited response time of the
data sources. We have shown that our solution and Linked Data can be applied
successfully in a real world scenario with signi cant advantages over established
Open Data solutions, e.g., a higher scalability, a more exible data schema and
rich queries with SPARQL. Since the European Commission (EC) establishes
DCAT-AP as a standard, our work acts as an enabler for its broad adoption.
Open Data from public administrations is an essential source for Linked Open
Data. Hence, our solution will have a direct impact on its dissemination. The
entire platform is available as Open Source.12
References
9 https://www.w3.org/TR/vocab-dqv/
10 https://www.docker.com/
11 https://kubernetes.io/
12 https://gitlab.com/european-data-portal</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          1.
          <string-name>
            <surname>Kirstein</surname>
            ,
            <given-names>F.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Dittwald</surname>
            ,
            <given-names>B.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Dutkowski</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Glikman</surname>
            ,
            <given-names>Y.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Schimmler</surname>
            ,
            <given-names>S.</given-names>
          </string-name>
          ,
          <string-name>
            <surname>Hauswirth</surname>
            ,
            <given-names>M.</given-names>
          </string-name>
          :
          <article-title>Linked Data in the European Data Portal: A Comprehensive Platform for Applying DCAT-AP</article-title>
          . In: EGOV2019 {
          <article-title>Joint Conference EGOV-CeDEM-EPART</article-title>
          <year>2019</year>
          (
          <year>2019</year>
          )
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          2.
          <string-name>
            <surname>Dragan</surname>
            ,
            <given-names>A.</given-names>
          </string-name>
          :
          <article-title>DCAT Application Pro le for data portals in Europe (</article-title>
          <year>2019</year>
          ), https://joinup.ec.europa.eu/sites/default/files/distribution/access_ url/2019-05/e3f7bcdf-eaad-4741
          <source>-9bf6-dc61327f4eea/DCAT_AP_1.2</source>
          .1.pdf
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>