Synopsis
Machine Learning for Named Entity Recognition
Description
The SEER project was intended to extend and generalise the state-of-the-art in entity recognition so that it could become a reliable enabling technology for a wide range of applications and higher-level NLP tasks. Our goal was to develop the means to recognize and classify a much wider range of entities than are traditionally treated, and to develop techniques which can be applied in a wide range of text types.
Rapid adaptation to new domains was a priority and this required modularity in the processing pipeline as well as the innovative use of machine-learning techniques to enable the use of unnannotated corpora as training material wherever possible.
People
University of Edinburgh
|
|
Start Date:
01 January 2002
End Date:
30 December 2004
|