* Apologies for cross-posting *
This special issue of the Journal of Web Semantics focuses on the
problem of finding links between datasets published as linked data.
Today the web of data has become a reality. The ever increasing number
of datasets published as RDF according to the linked data principles,
the support of major search engines, e-commerce sites and social
networks give no doubt that the early scenarios of the semantic Web
vision will soon become a reality.
The power of the web lies in its networked structure, in the connections
between the resources it contains. Similarly, linked data enable the
interlinking of data resources so that databases become interconnected
and the information they contain become part of a huge distributed
database. The transformation of the Web from a “Web of documents” into a
“Web of data”, as well as the availability of large collections of
sensor generated data (“internet of things”), is leading to a new
generation of Web applications based on the integration of both data and
services. At the same time, new data are published every day out of user
generated contents and public Web sites.
This emergence of the Web of data raises many challenges, such as the
need of comparing and matching data with the goal of resolving the
multiplicity of data references to the same real-world objects and of
finding useful and relevant similarities and correspondences among data.
The Web needs techniques and tools for the discovery of data links, and
a suitable theory for the understanding and definition of the data links
meaning.
About data links, one of the most important goals is to provide means to
ensure that the interconnection between data is effective. The design of
algorithms, methodologies, languages and tools that provide more
efficient and automated ways to link data is essential for the growth of
the Web of data rather than a set of disjoint data islands.
While the problems of entity resolution have been studied in the
database community for a long time, the Web of data environment presents
new important challenges at different levels. Large volumes of data and
the variety of repositories which have to be processed rise the need for
scalable linking techniques which require minimal user involvement. On
the other hand, in cases where user configuration effort is required,
there is a need for tools to be usable by non-experts in the domain.
Given that published data links can be used by automatic reasoning
tools, it is important to capture the meaning of links in a precise way.
Since quality of automatically generated links can vary, their
provenance and reliability have to be modelled in an explicit way.
Finally, to capture and compare the reliability of different tools and
techniques, there is a need for evaluation methods for automatic data
linking approaches.
Challenges
• Automating the process of finding links between Web datasets
• Scaling data linking algorithms
• Representation and interpretation of links
• Providing efficient user interfaces and interaction methods
• Modeling and reasoning on links trust and provenance information
Topics of Interest
The topics of interest for this special issue include but are not
limited to the following.
• data linking tools and frameworks
• techniques for automated data linking
• data similarity measures
• similarity spreading measures
• schema-based similarity measures
• candidate dataset selection and datasets similarity measures
• statistical analysis techniques
• semi-supervised, learning-based data linking methods
• optimization methods for computing similarity
• web data sampling techniques
• identity representation and semantics
• reasoning on links, link propagation
• user interaction for link elicitation and validation
• provenance and trust models on links
• methods for link quality assessment
• innovative applications using links
• evaluation of data linking techniques and tools
Important Dates
We will review papers on a rolling basis as they are submitted and
explicitly encourage submissions well before the final deadline.
• 1 June: submission deadline
• 1 September: initial decisions and notifications
• 1 October: major/minor revisions due
• 1 November: final minor revisions due
• 1 December: final decisions and notifications
• 1 January: preprints available publication in 2013
Instructions for submission
Please see the author guidelines for detailed instructions before you
submit. Submissions should be conducted through Elsevier’s Electronic
Submission System. More details on the Journal of Web Semantics can be
found on its homepage. See the JWS Guide for Authors for details on the
submission process.
Editors
• Alfio Ferrara (Università degli Studi di Milano)
• Andriy Nikolov (Open University)
• François Scharffe (LIRMM, Université de Montpellier 2)
