Framework for construction and incremental maintenance of high-quality linked data mashup

Abstract

Due to the Linked Data initiative, previously isolated datasets are published as linked data. This enables the creation of applications that consume data from multiple Linked Data sources. Applications are confronted with the challenge of obtaining a homogenized view of this global data space, called a Linked Data Mashup view. This work proposes a framework to perform the fusion of Linked Data and quality assessment of Linked Data Mashup. Quality assessment of Linked Data mashup is computed based on the result of the data fusion. We also propose the implementation of a platform for creation and incremental maintenance of mashup views.