Linked data provenance: state of the art and challenges
Sarawat Anam, Byeong Ho Kang, Yang Sok Kim, Qing Liu · eCite Digital Repository (University of Tasmania) · 2015
Linked Open Data (LOD) is rapidly emerging inpublishing and sharing structured data over the semanticweb using URIs and RDF in many application domainssuch as fisheries, health, environment, education andagriculture. Since different schemas that have the samesemantics are found in different datasets of the LODCloud, the problem of managing semantic heterogeneityamong the schemas is increasing. Schema level mappingamong the datasets of the LOD Cloud is necessary asinstance level mapping among the datasets is not feasiblein the process of making knowledge discovery easy andsystematic. In order to correctly interpret query resultsover the integrated dataset, schema level mappingprovenance is necessary. In this paper, we review existingapproaches of linked data provenance representation,storage and querying, and applications of linked dataprovenance where mapping is at the instance level. Theanalysis of existing approaches will assist us in revealingopen research problems in the area of linked dataprovenance where mapping is at the schema level.Furthermore, we explain how schema level mappingprovenance in linked data can be used to facilitate dataintegration and data mining, and also to ensure qualityand trust in data.