About
The airpedia project aims to extract structured information from Wikipedia (and other unstructured or semi-structured sources) and make it freely available. One can query the dataset using our SPARQL endpoint. The project is mainly based on DBpedia, that uses Wikipedia infoboxes to extract information. Our approach uses a machine learning algorithm that, starting from pages included in DBpedia, trains a classifier that extracts new instances of pages when the infobox is not present.
For now, the resource provides only the extended population of the DBpedia ontology. In the future we will also populate properties. We will update on news through our blog.
You can download the available resources from our download section.
Latest blog posts
- 25/01/2015 Source code released
- 30/09/2014 Esperanto DBpedia released
- 06/03/2014 Paper accepted at ESWC 2014
- 13/09/2013 Paper accepted at NLP & DBpedia workshop (ISWC)
- 04/07/2013 Paper accepted at ISWC conference