Back to Home - Daniel Alexander Smith

Linked Data Engine: A selection of open source tools to aid those wishing to participate in and take advantage of the linked data ecosystem. Specifically the following discrete tools are available here:

• Data Processor: A lightweight web management interface for adding knowledge bases to triplestores and adding RDF sources to those triplestores. Written as PHP5 objects using an MVC architecture. The triplestore backend is genericised, loading appropriate backend connectors dynamically, and includes a backend connector for OpenLink Virtuoso. All operations (list KBs, add new KB, get KB detail and add RDF source to KB) are possible using a human UI as well as a RESTful XML API.

  The purpose of the Data Processor is to provide a stable web API to interact with SPARQL endpoints, regardless of the specific triplestore software behind the endpoint. The core operations supported are as follows:

  • Add KB: Add a new Knowledge Base (optionally specify the name of the KB, otherwise a UID will be generated).
  • Add a source to a KB: Add a specified RDF resource or file (specified by URI) to a specified KB.
  • List KBs: Returns the IDs of the KBs in the system.
  • KB details: Specify the ID of a KB and it returns the URIs of the sources in a KB, and the URL to the SPARQL endpoint of the KB.

  Data Processor is used by my web applications such as Data Picker, which natively query SPARQL, but need a way to query RDF over the web, or from Linked Data. For an alternative to this approach, see Tabulator from Professor Sir Tim Berners-Lee's team at MIT, where a local JavaScript triplestore gathers RDF from the web, and queries it using SPARQL (it's magic and scary, but not in the creepy clown sense).

  The core code to the data processor will be uploaded here and additionally hosted on github. It will likely be licensed under the BSD License or the MIT License.