Improving Natural Language Processing by Linguistic Document Annotation
Hideo Watanabe, Katashi Nagao, Michael C. McCord, Arendse Bernth · International Conference on Computational Linguistics · 2000
Natural language processing (NLP) programs are confronted with various difficulties in processing HTML and XML documents, and have the potential to produce better results if linguistic information is annotated in source texts. We have therefore developed the Linguistic Annotation Language (or LAL), which is an XML-compliant tag set for assisting natural language processing programs. It consists of linguistic information tags such as tags specifying word/phrasal boundaries, and task-dependent instruction tags such as tags defining the scope of translation for machine translation programs. We have also developed an LAL-annotation editor to facilitate users to annotate documents without seeing tags.