Paper: Towards Morphologically Annotated Corpus of Hospital Discharge Reports in Polish

ACL ID W11-0211
Title Towards Morphologically Annotated Corpus of Hospital Discharge Reports in Polish
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2011
Authors

The paper discuses problems in annotating a corpus containing Polish clinical data with low level linguistic information. We propose an approach to tokenization and automatic morphologic annotation of data that uses ex- isting programs combined with a set of do- main specific rules and vocabulary. Finally we present the results of manual verification of the annotation for a subset of data.