The Open NEE Configuration Model is an RDF/S vocabulary for describing a configuration supported by a Named Entity Extraction (NEE) service. It is based on the approach proposed in "Configuring Named Entity Extraction through Real-Time Exploitation of Linked Data" (WIMS'14) for configuring such services, and allows a NEE service to describe and publish as Linked Data its entity mining capabilities, but also to be dynamically configured.


Adopting an open and exchangeable configuration model offers many benefits including:

  • Exchangeability and Portability. Configurations can be exchanged by users/communities, e.g. for annotating different corpora of documents using the same configuration, i.e. the same categories, lists of entities, KBs, etc. In addition, the availability of a model like the one that we propose enables a NEE service to offer an API that accepts and uses such configurations, while the result of the annotation process can be in a standard format, allowing its further exploitation in several contexts.
  • Aggregation and Integration of multiple configurations. A common model allows someone to collect such configurations (provided by different NEE systems) and then, by querying them, to select those systems that satisfy the needs of the intended application.
  • Benchmarking. Common configurations would allow comparative evaluation of different NEE systems, e.g. with respect to efficiency, effectiveness of entity disambiguation, etc.
  • Extendability. The expression of the model as an RDF Schema allows someone to extend it by exploiting also other vocabularies.

releases top
examples top
publications top
P. Fafalios, M. Baritakis and Y. Tzitzikas
Exploiting Linked Data for Open and Configurable Named Entity Extraction,
International Journal on Artificial Intelligence Tools, Vol. 24, No. 02 (2015), World Scientific (ISSN: 1793-6349).
contact / provide feedback top
Please provide your feedback and any comments by sending an email at tzitzik at ics.forth.gr.
Institute of Computer Science - Foundation for Research and Technology - Hellas
Terms of Use  |   Privacy Policy