RDF Generation and Linking using Open Refine Daniel

RDF Generation and Linking using Open Refine Daniel Vila-Suero dvila@fi. upm. es

Introduction • Learn how to use Open Refine and its RDF extension • Map tabular data to RDF • Link entities to a external dataset

Overview 3

Installation • Download Open Refine at: • http: //openrefine. org/download. html • Download the RDF Extension at: • http: //refine. deri. ie/

Installing the RDF Extension • Make sure "extensions" folder exists in your Google Refine workspace • Download the extension • Extract the downloaded zip file to the "extensions" folder • Start/Restart Google Refine

Data • Download available at: http: //data. freeyourmetadata. org/powerhousemuseum/phm-collection-cleaned. zip

Task 1 • Inspect the data • Pre-process if neccesary

Task 2 • Using the RDF extension map the data to RDF using a selection of vocabularies

Task 3 • Using a reconciliation service try to find entities within your data that have URIs in other dataset. • For example the LCSH (Library of Congress Subject Heading) with the following configuration: • Name: LCSH (preprocessed) • Endpoint URL: http: //sparql. freeyourmetadata. org/ • Graph URI: http: //sparql. freeyourmetadata. org/authoritiesprocessed/ • Type: Virtuoso • Label properties: check only skos: pref. Label

Publish • Load the data in a SPARQL endpoint (Virtuoso, 4 store, Fuseki, Sesame, etc. ) • Install and configure a LD front-end: • Recommended: Puelia (PHP), Elda (Java) • Configurable using a config file in turtle • Documentation available at: https: //code. google. com/p/linked-dataapi/wiki/Specification 10