The project developed three demonstrators, that is, three SemaGrow Stack client applications:
- The Trees4Future/AgMIP Demonstrator validates the Semagrow system on the Heterogeneous Data Collections & Streams use cases. This demonstrator focuses on the support of scientific communities in the domain of agricultural and forestry modelling. Specifically, it supports end-users in preparing suitable input dataset for their modelling exercises from the wealth of heterogeneous big data collections and streams available, validating the integrative semantic capabilities provided by the Semagrow Stack as well as its ability to search and retrieve large data volumes.
- The AGRIS Demonstrator validates the SemaGrow system on the Reactive Data Analysis use cases. The AGRIS Web Portal serves bibliographical data from the agricultural domain, combined with widgets that retrieve and present Web resources that are relevant to the currently viewed bibliography item. The AGRIS Demonstrator validates that using the SemaGrow Stack as a backend supports IT personnel more easily extend the range of data sources federates under the AGRIS Web Portal; and that the Semagrow Stack is able to reactively search through big data in order to find results that are not voluminous.
- The Linked Open Data Hub (LODH) application validates the SemaGrow system on the Reactive Resource Discovery use cases. LODH provides the user with the ability to retrieve of resources that are related with the resources of the hub but they are from other data sources in the SemaGrow Stack, such as educational material, academic publications, lectures, and presentations. The LODH Demonstrator validates Semagrow technologies on Web developers with simpler, NoSQL querying use cases and for use cases where end-users need to search a large number of data sources, although the contents of any individual data source do not constitute big data by itself.
In order to achieve these deployments, the project prepared and provisioned hardware infrastructure and developed software for accessing it, server-side tools, clients and client-side tools, seeding an ecosystem of software tools, semantic vocabularies, and public Web services developed around the main SemaGrow system.
In particular, the Trees4Future/AgMIP Demonstrator makes a heavy use of the rdfCDFToolkit to convert NetCDF files to the RDF format, the AGRIS Demonstrator use the cleanDT tool to clean the dates of the AGRIS dataset and the Linked Open Data Hub (LODH) pilot use the SemagrowREST tool to encapsulate the SPARQL endpoint inside a REST API. All the pilots use the authoring tools ELEON and metadatagen in order to generate appropriate data source descriptions for their federation. A virtual machine with a complete and fully configured deployment, exactly as used during piloting, can be found in the Tools section.