Building a Large Annotated Corpus of English: The Penn Treebank

1 Introduction

1. What is the Penn Treebank, as described in 1993?
   A corpus of over 4.5 million words of American English annotated for part of speech, with over half also annotated for skeletal syntactic structure, built 1989–1992 and distributed through the Linguistic Data Consortium.
   • Marcus et al., 1993, Building a Large Annotated Corpus of English (Computational Linguistics), §1
2. What uses did the authors foresee for large annotated corpora?
   Automatically building statistical models of grammar for written and spoken language, formal theories of the grammars of writing and speech, studying prosody, and evaluating and comparing parsing models.
   • Marcus et al., 1993, Building a Large Annotated Corpus of English (Computational Linguistics), §1
3. What is the general annotation method used for both tagging and bracketing in the Penn Treebank?
   Hand correction of the output of an errorful automatic process: a stochastic tagger, then a parser, produce a first pass that annotators correct.
   • Marcus et al., 1993, Building a Large Annotated Corpus of English (Computational Linguistics), §1