Web of Trusted Data - Trustworthiness of Data described by RDF

Today, large amounts of RDF data are published on the Web; large datasets are interlinked; new applications emerge that utilize this data in novel and innovative ways. However, the openness of the Web and the ease to combine RDF data from different sources creates new challenges. Unreliable data could dominate the result of queries, taint inferred data, affect local knowledge bases, or may have negative or misleading impact on software agents. Hence, questions of reliability and trustworthiness must be addressed. While several approaches consider trustworthiness of information sources, little has been done considering the actual data itself. More precisely, what is missing for the Web of data is a uniform approach to rate the trustworthiness of the data on the Web and standardized mechanisms to access and to use those ratings.

With our work we address these open issues; we aim for an evolution of the Web of data to a "Web of trusted data." As a basis we develop a trust model for RDF that must be adequate as well as simple and reasonable to gain wide-spread use. The envisioned Web of trusted data builds upon this model; all data must be rated accordingly. Thus, we investigate strategies how to rate and how to determine the trustworthiness of RDF data. Those strategies may consider the origin of statements as well as the opinion of other consumers of the data.

Naturally, we do not consider the plain existence of trustworthiness ratings as an end in itself; users as well as software agents have to be able to utilize the trust ratings and base their decisions upon them. Hence, we investigate usage scenarios and possible applications. A main requirement is the ability to access the trustworthiness ratings. Thus, we study data access methods in the Web of data and develop concepts that extend those methods accordingly. For instance, we consider trust-aware extensions for the RDF query language SPARQL.

A realization of the Web of trusted data requires trust management components that implement our ideas. Based on an in-depth analysis of the requirements for these components we propose possible solutions with respect to efficient representation and processing of the trustworthiness ratings.

Publications

Trustworthiness of Data on the Web
Olaf Hartig
Proceedings of the STI Berlin & CSW PhD Workshop, Berlin, Germany, 2008/09 (pdf)

Querying Trust in RDF Data with tSPARQL (Abstract)
Olaf Hartig
Proceedings of the 6th European Semantic Web Conference (ESWC), Heraklion, Greece; (Best Paper Award), 2009/06 (pdf)

Provenance Information in the Web of Data
Olaf Hartig
Proceedings of the Linked Data on the Web (LDOW'09) Workshop at the World Wide Web Conference (WWW), Madrid, Spain, 2009/04 (pdf)

Using Web Data Provenance for Quality Assessment
Olaf Hartig, Jun Zhao
Proceedings of the 1st International Workshop on the Role of Semantic Web in Provenance Management (SWPM) at the International Semantic Web Conference, 2009/10 (pdf)

Integrating Provenance into the Web of Data
Olaf Hartig, Jun Zhao
Proceedings of the Poster Session at the 7th Extended Semantic Web Conference (ESWC), Heraklion, Greece, 2010/06 (pdf)

Towards a Data-Centric Notion of Trust in the Semantic Web (A Position Statement)
Olaf Hartig
Proceedings of the 2nd Workshop on Trust and Privacy on the Social and Semantic Web (SPOT) at the 7th Extended Semantic Web Conference (ESWC), Herakli, 2010/06 (pdf)

Publishing and Consuming Provenance Metadata on the Web of Linked Data
Olaf Hartig, Jun Zhao
Proceedings of the 3rd International Provenance and Annotation Workshop (IPAW), Troy, New York, USA, 2010/06 (pdf)

Towards Interoperable Provenance Publication on the Linked Data Web
Jun Zhao, Olaf Hartig
Proceedings of the 5th Linked Data on the Web (LDOW) Workshop at the World Wide Web Conference (WWW), Lyon, France, 2012/04 (pdf)