Linked widgets platform for rapid collaborative semantic mashup development

In recent years, data has become vital in supporting our everyday lives. Along with large volumes of open data available on the web, various types of public, private, and enterprise data are stored in the cloud or distributed over multiple devices. The value of this data would increase drastically if we were able to integrate it. This would enable more sophisticated presentation and analysis of previously disparate data. So far, however, it is challenging for non-expert users to efficiently make use of such data because (i) data heterogeneity hampers integration of different kinds of data that are stored in various formats and spread among storage infrastructures; (ii) manual data integration processes are typically neither reproducible, nor reusable; and (iii) the lack of support for exploration does not allow for the integration of arbitrary data sources. This paper tackles these challenges by introducing a mashup platform that combines semantic web and mashup concepts to help users obtain insights and make informed decisions. To this end, we leverage a semantic model of mashup components for automated techniques that support the user in exploring available data. Moreover, we introduce a collaborative and distributed model to create and execute mashups. This facilitates distributed ad-hoc integration of heterogeneous data contributed by multiple stakeholders.