This joint volume of proceedings gathers together papers from the 2nd Workshop on Managing the Evolution and Preservation of the Data Web (MEPDaW) and the 3rd Workshop on Linked Data Quality (LDQ), held on the 30th of May of 2016 during the 13th ESWC conference in Anissaras, Crete, Greece. Managing the Evolution and Preservation of the Data Web This workshop targeted one of the emerging and fundamental problems in the Semantic Web, specifically the preservation of evolving linked datasets. There is a vast and rapidly increasing quantity of scientific, corporate, government and crowd-sourced data published on the emerging Data Web. Open Data are expected to play a catalyst role in the way structured information is exploited in the large scale. This offers a great potential for building innovative products and services that create new value from already collected data. It is expected to foster active citizenship (e.g., around the topics of journalism, greenhouse gas emissions, food supply-chains, smart mobility, etc.) and world-wide research according to the “fourth paradigm of science”. The most noteworthy advantage of the Data Web is that, rather than documents, facts are recorded, which become the basis for discovering new knowledge that is not contained in any individual source, and solving problems that were not originally anticipated. In particular, Open Data published according to the Linked Data Paradigm are essentially transforming the Web into a vibrant information ecosystem. Published datasets are openly available on the Web. A traditional view of digitally preserving them by “pickling them and locking them away” for future use, like groceries, would conflict with their evolution. There are a number of approaches and frameworks, such as the LOD2 stack, that manage a full life-cycle of the Data Web.
? MEPDaW+LDQ join proceedings are publicly available in [
Preserving linked open datasets poses a number of challenges, mainly related to the nature of the LOD principles and the RDF data model. In LOD, datasets representing real-world entities are structured; thus, when managing and representing facts we need to take into consideration possible constraints that may hold. Since resources might be interlinked, effective citation measures are required to be in place to enable, for example, the ranking of datasets according to their measured quality. Another challenge is to determine the consequences that changes to one LOD dataset may have to other datasets linked to it. The distributed nature of LOD datasets furthermore makes archiving a headache.
This workshop aimed at addressing the above mentioned challenges and issues by providing a forum for researchers and practitioners who apply linked data technologies to discuss, exchange and disseminate their work.
The workshop included an inspiring talk by Dr. Axel Polleres on Archiving Linked and Open Data, three research papers and one industry paper, and a plenary discussion. Based on the review scores, the best paper award has been given to Ruben Taelman, Ruben Verborgh, Pieter Colpaert, Erik Mannens and Rik Van de Walle for their work “Continuously Updating Query Results over Real-Time Linked Data”. 2
The focus of this workshop was to reveal novel methodologies and frameworks in assessing, monitoring, maintaining, and improving the quality of Linked Data (LD) as well as introduce tools and user interfaces which can effectively assist in the assessment and repair. In addition, the workshop sought methodologies that help to identify the current impediments in building real-world Linked Data applications leveraging data quality. The benefits of addressing Linked Data quality issues would not only help in detecting inherent data quality problems currently plaguing Linked Data, but also provide the means to fix these problems and maintain the quality in the long run.
To guarantee the full exploitation of the published or consumed Linked Data, it is important to assure LD quality. In this way, it is possible to understand whether data is appropriate for the task at hand before using it. There are several issues in LD which hampers the use of datasets in building real-world LD-based applications and research solutions. One of issues is the method to find the most relevant LD data for a particular application. Also, generating meaningful associations between the LD datasets, at the ontology, data or property level is an important issue to be considered when building such applications. Essentially, the quality of LD is a deciding factor as to which datasets can be used in building such real-world applications. Currently, there is no full-proof method of performing this kind of quality assessment.
In general detecting the quality of datasets available and making this information explicit is a challenge. This entails the (semi-)automatic identification of existing problems, which is either insensitive to the use case or is limited in identifying only specific (objective) quality issues. In LD few efforts are currently available to standardize how data quality tracking and assurance should be implemented and it poses yet other challenges: (i) LD refers to a Web-scale knowledge base consisting of interlinked published data from a multitude of autonomous information providers (variety of data). The quality of provided information may depend on the intention of the information provider; (ii) the increasing diffusion of the LD paradigm as a standard way to share knowledge on the Web allows consumers to fully exploit vast amount of data that were not available in the past (high volume of data). We are likely to find more low quality in LD than in smaller data sets because in large data sets data are produced with automatic information processes which are often error prone; (iii) data sets in LD formats may often be used by third-party applications in ways not expected by the original creators of the data set; (iv) LD provides data integration through interlinking of data between heterogeneous data sources. The quality of integrated data will depend on the quality of original data sources; (v) last but not least relevant, Linked Data can be considered as a dynamic environment where information can change rapidly and cannot be assumed to be static (velocity of data). Changes in LD sources should reflect changes in the real world, otherwise data can soon become out-dated. Out-of-date information can reflect data inaccuracy problems and can deliver invalid information. For example, more upto-date information should Data be preferred over less up-to-date information in data integration and fusion applications. Moreover, none of the current approaches use the assessment to ultimately improve the quality of the underlying dataset.
The workshop included a keynote by Christian Dierschl on Data quality assurance
in data-intensive systems, three paper presentations and a lightning talk session with
following discussions. Based on the review scores, the best paper award has been given
to Toma´sˇ Knap for his work on Increasing Quality of Austrian Open Data by Linking
them to Linked Data Sources: Lessons Learned [
Organising Committees – MEPDaW
Jeremy Debattista, Enterprise Information Systems, University of Bonn, Germany / Organized Knowledge, Fraunhofer IAIS, Germany Ju¨rgen Umbrich, Vienna University of Economics and Business, Austria Javier D. Ferna´ndez, Vienna University of Economics and Business, Austria – LDQ
We would like to thank the authors for their contribution and active participation in the workshops, and all the program committee members for reviewing the submissions and provide valuable feedback. We are also grateful to the organisers of the ESWC 2016 conference for their support, and our keynote speakers, Axel Polleres from the Vienna University of Economics and Business (Austria), and Christian Dirschl from Wolters Kluwer (Germany).
The MEPDaW workshop was co-organised by members funded by the Austrian Science Fund (FWF): M1720-G11.
The LDQ workshop was co-organised by members funded by the German Government, Federal Ministry of Education and Research under the project: 03WKCJ4D.