Publishing E-RDF linked data for many agents by single third-party server

TY - GEN

T1 - Publishing E-RDF linked data for many agents by single third-party server

AU - Wang, Dongsheng

AU - Zhang, Yongyuan

AU - Wang, Zhengjun

AU - Chen, Tao

N1 - Conference code: 7

PY - 2017

Y1 - 2017

N2 - Linked data is one of the most successful practices in semantic web, which has led to the opening and interlinking of data. Though many agents (mostly academic organizations and government) have published a large amount of linked data, numerous agents such as private companies and industries either do not have the ability or do not want to make an additional effort to publish linked data. Thus, for agents who are willing to open part of their data but do not want to make an effort, the task can be undertaken by a professional third-party server (together with professional experts) that publishes linked data for these agents. Consequently, when a single third-party server is on behalf of multiple agents, it is also responsible to organize these multiple-source URIs (data) in a systematic way to make them referable, satisfying the 4-star data principles, as well as protect the confidential data of these agents. In this paper, we propose a framework to leverage these challenges and design a URI standard based on our proposed E-RDF, which extends and optimizes the existing 5-star linked data principles. Also, we introduce a customized data filtering mechanism to protect the confidential data. For validation, we implement a prototype system as a third-party server that publishes linked data for a number of agents. It demonstrates well-organized 5-star linked data plus E-RDF and shows the additional advantages of data integration and interlinking among agents.

AB - Linked data is one of the most successful practices in semantic web, which has led to the opening and interlinking of data. Though many agents (mostly academic organizations and government) have published a large amount of linked data, numerous agents such as private companies and industries either do not have the ability or do not want to make an additional effort to publish linked data. Thus, for agents who are willing to open part of their data but do not want to make an effort, the task can be undertaken by a professional third-party server (together with professional experts) that publishes linked data for these agents. Consequently, when a single third-party server is on behalf of multiple agents, it is also responsible to organize these multiple-source URIs (data) in a systematic way to make them referable, satisfying the 4-star data principles, as well as protect the confidential data of these agents. In this paper, we propose a framework to leverage these challenges and design a URI standard based on our proposed E-RDF, which extends and optimizes the existing 5-star linked data principles. Also, we introduce a customized data filtering mechanism to protect the confidential data. For validation, we implement a prototype system as a third-party server that publishes linked data for a number of agents. It demonstrates well-organized 5-star linked data plus E-RDF and shows the additional advantages of data integration and interlinking among agents.

KW - Data integration

KW - E-RDF

KW - Knowledge representation

KW - Linked data

KW - Semantic web

KW - Web service

UR - http://www.scopus.com/inward/record.url?scp=85033791249&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-70682-5_10

DO - 10.1007/978-3-319-70682-5_10

M3 - Article in proceedings

AN - SCOPUS:85033791249

SN - 978-3-319-70681-8

T3 - Lecture notes in computer science

SP - 151

EP - 163

BT - Semantic Technology

A2 - Wang, Zhe

A2 - Turhan, Anni-Yasmin

A2 - Wang, Kewen

A2 - Zhang, Xiaowang

PB - Springer

T2 - 7th Joint International Conference on Semantic Technology

Y2 - 10 November 2017 through 12 November 2017

ER -