Contents:
- 1. Introduction
- 2. Converting Structured Data to RDF
- 3. Adding Text Relations – txt2rdf
- 4. Query Experiments
- 5. Conclusions
- References
This paper explains how to augment structured data with “facts” automatically extracted from free text. This is done via a pipeline, txt2rdf, which takes in plain text at one end and pumps out RDF triples at the other. These triples are then combined with related Semantic Web data. The author acknowledges that the process is not yet perfect, but the routes to improve it are very clear. Two experiments have shown the pipeline to produce an integrated RDF graph structure that can answer queries for information that was impossible to retrieve previously.
Clear paper that proves in practice the potential for powerful and flexible public access systems to cultural (AV-)archives everywhere. The described txt2rdf pipeline gives AV-archivists the opportunity to choose which textual context information might best enrich their catalogue descriptions (linked data).