COLD 2014 - Fifth International Workshop on Consuming Linked Data (COLD2014)

The term Linked Data refers to a set of foundational principles for publishing and interlinking structured data on the Web. After Linked Data was first proposed in 2006, a grass-roots movement, led by the Linking Open Data project, started to publish and to interlink multiple open databases on the Web following the proposed principles. Due to conference workshops, tutorials and general evangelism, an increasing number of data publishers ? such as the BBC, Thomson Reuters, The New York Times, the Library of Congress, BestBuy, Getty, the US and UK government ? have since adopted this practice. This ongoing effort resulted in bootstrapping the “Web of Linked Data” which, today, comprises of billions of RDF triples and millions of RDF links between datasets. The published datasets now include data about books, movies, music, radio and television programs, reviews, scientific publications, genes, proteins, diseases, medicine and clinical trials, geographic locations, people, statistical and census data, companies, and many more topics besides.
All of these published datasets are openly available on the Web in standardised interoperable formats, which presents novel opportunities for the next generation of Web-based applications: data from different providers can be aggregated, allowing fragmentary information from multiple sources to be integrated so as to achieve a complementary and more complete view. While a few applications, such as the BBC music guide have used Linked Data to significant benefit, the deployment methodology has been to harvest the data of interest from the Web to create a private, disconnected repository for each specific application. Such “harvesting approaches” are typically only feasible for vertical applications tied to specific datasets, incur a high up-front cost, and are insensitive to updates in the original data-sources. New concepts for consuming Linked Data ? that do not require up-front harvesting of all sources ? are required to lead the Web of Linked Data to its fullest and most general potential. The concepts, patterns, and tools necessary are very different from situations where relevant resource identifiers are known a priori, where queries can be run over complete local repositories, where access to the repository is reliable and cheap, and where relevant data sources are known to be trustworthy.
While recent tutorials provided technical insight and skills on how to consume Linked Data in applications, these tutorials also drew attention to several open issues that make the development of Linked Data based applications a challenging or still impossible task. These open issues include a lack of approaches for seamless integration of Linked Data from multiple sources, for dynamic, on-the-fly discovery of available data, for information quality assessment, for querying and caching dynamic remote sources, and for implementing appropriate end-user interfaces.
These open issues can only be addressed appropriately when they are conceived as research problems that require the development and systematic investigation of novel approaches. The 5th International Workshop on Consuming Linked Data (COLD 2014) aims to provide a platform for the presentation and discussion of such approaches. Our main objective is to attract submissions that present scientific discussion (including systematic evaluation) of broadly-applicable concepts and approaches, as opposed to exposition of features implemented in current Linked Data applications. For practical systems without formalization or evaluation we refer interested participants to other offerings at ISWC, such as the Semantic Web Challenge or the Demo Track. As such, we see our workshop as orthogonal to these events.
Topics of Interest
While previous editions of the workshop have attracted a number of submissions that addressed topics related to (RDF and) Linked Data management in general, with COLD2014 we aim to steer the workshop back towards the aforementioned core goals. To this end, we explicitly seek submissions that address research problems related to at least one of the following two aspects of Linked Data consumption:
Makes use of Linked Data principles, including dereferencing
Involves direct use of multiple, real-world Linked Datasets
In the context of these two aspects of Linked Data consumption, relevant topics for COLD 2014 include but are not limited to:
Live Linked Data (i.e., algorithms and applications that make use of Linked Data at runtime)
Architectures for consuming Linked Data (e.g., Dataspaces, Cloud, NoSQL)
Integration of Linked Data sources (e.g., entity resolution, sameas, vocabulary mapping, etc.)
Handling additional Web data (e.g., Deep Web, APIs, Microdata, JSON, Atom, tables, etc.)
Web-scale data management (e.g., crawling, indexing, parallel processing, etc.)
Novel languages for navigating and consuming Linked Data (e.g., nSPARQL, NautiLOD, etc)
Linked Data summarization, guides and schema learning
Query processing over multiple Linked Datasets
Search over the Web of Linked Data
Auto-discovery of URIs and data
Caching and replication
Dataset dynamics
Reasoning on Linked Data from multiple sources
Information quality and trustworthiness of Linked Data
User-interface research for interacting with the Web of Linked Data