Monitoring the dynamism of the linked data space through environment abstraction

Abstract

Management of data dynamics is a major issue for linked data applications. There are some approaches and applications addressing this problem which are either pull or push based. However, these one sided solutions do not solve the problem since pull based approaches bring an overhead to data consumers, and push based approaches require the collaboration of the data publishers which is contrary to the free nature of the Web of Data. In this paper, to monitor the dynamism of the linked data space, we present an infrastructure that combines pull and push approaches to monitor the changes of SPARQL queries over the linked data space. In the design of this infrastructure, environment abstraction of multi-agent systems domain is followed. The proposed infrastructure has been built upon the CArtAgO environment infrastructure, and VoID based federated query engine, WoDQA, is incorporated into CArtAgO to handle dynamic discovery of datasets and execution of monitored SPARQL queries.

Data Management and Ontology Development for Provenance-Aware Organizations in Linked Data Space

“…The metadata of datasets relevant to the organization should be stored locally within the organization, referred to as local VoID metadata. Organizational Data Source Metadata should be able to keep up-to-date by tracking changes in Global Data Source Metadata [29,30]. This allows the organization to access the data it needs in a way that meets its expectations in a current form.…”