iDocument: Intelligent Document Information Extraction

iDocument is a generic ontology-based information extraction (OBIE) system that uses ontological background knowledge in terms of existing vocabularies and instance knowledge. iDocument uses existing knowledge from personal or business domains (e.g. relational databases, concept maps, taxonomies, etc.). Following Semantic Web, iDocument exchanges and extracts knowledge based on the W3C standard RDF. Existing knowledge is used as input in a serial IE pipeline of extraction tasks for extracting possible answers concerning user specified ad hoc queries on a given text collection.

Projects

• iDocument was started in project ​iDocument in 2008. Project iDocument was funded by ​Stiftung Rheinland-Pfalz für Innovation
• Further developement and research is done in project ​Perspecting. Perspecting is funded by the Bundesministerium für Bildung und Forschung (Federal Ministry of Education and Research) under grant 01IW08002.

Unique Feature

• Domain ontologies are exchangeable as long as they are written in RDFS.
• The MOBIE mapping vocabulary allows to define relevant classes, attributes and relations for extraction purpose .
• Existing instance knowledge is reused for information extraction purpose
• Extracted results are formalized in the same RDF scheme as the input domain ontology.
• SPARQL queries are used for defining extraction templates.
• All intermediate and final extraction results are weighted hypothesis according to Dempster –Shafer’s belief function.

Table of Contents

• System Summary
• DFKI Olympic Corpus and Annotation Scheme 2008 (DFKI OCAS 2008)
• ​Publications(1)
• ​Publications(2)
• Reference Projects
• ​Poster
• ​Flyer

Contact

For further information please contact ​Benjamin Adrian (benjamin.adrian@…).

This project is developed at ​DFKI Knowledge Management Department