Corpus Name: English Chinese Translation Treebank, v. 1.0

Language(s):
English, translated from Chinese

Authors: 	
Ann Bies, Martha Palmer, Colin Warner, Justin Mott

Publication Type:
Text

Source Data Type:
Text

Source Data Genre:
Newswire

Project:
TIDES, NSF

Data Source Type:
Text

File Format(s):
Text

Size: 	
7.5 MB

Tokens:
146,300 words

Applications:
Natural language processing, parsing, tagging, machine translation

Description:

This release of the English Chinese Treebank consists of 146,300 words
in 325 files of individual Xinhua news stories (corresponding to the
Xinhua data in the Chinese Treebank 5.0, LDC Catalog No.: LDC2005T01)
that are translated into English, part-of-speech tagged and
treebanked.  The files were compressed using gzip.

The source files for the treebank annotation contain the final updated
translation of these files.  Translation errors that prevented
complete treebank annotation have been corrected.  This translation
and annotation were completed in October 2004, and this supersedes any
earlier translation.

The data can be found in the following directories:

  Penn Treebank-style files (converted from WordFreak annotation files)
  /data/pennTB-style-trees/

  Penn Treebank-style files (reformated and indented for easier human reading)
  /data/reformatted-PTBstyle/

  WordFreak annotated files (the files as they were annotated using the
			    Wordfreak annotation tool)
  /data/annotated-files/

  Source text files (corrected translations)
  /data/rawtext-files/

The guidelines followed for both part-of-speech and treebank annotation 
are essentially Penn Treebank II style, with two notable differences:

1. POS: tokenization of hyphenated items ("New York-based" has been
   replaced by "New York - based" for example), and the addition of HYPH
   and AFX tags necessitated by this change in tokenization

2. TreeBank: the addition of the node label NML for sub-NP nominal
   constituents (replacing NX and most NP-internal NAC)

More detailed addenda to the Penn Treebank II guidelines can be found in
/docs/pos-guidelines-addenda.txt and
/docs/treebank-guidelines-addenda.txt

The Penn Treebank II style guidelines (Bracketing Guidelines for
Treebank II Style, Eds: Ann Bies, Mark Ferguson, Karen Katz, Robert
MacIntyre, Penn Treebank Project, University of Pennsylvania, CIS
Technical Report MS-CIS-95-06, 1995) and the Penn Treebank
part-of-speech tagging guidelines are available at
http://www.cis.upenn.edu/~treebank/

The Wordfreak annotation tool is available at http://wordfreak.sourceforge.net/

Project manager: 	
Ann Bies, bies@ldc.upenn.edu

Annotators:
Justin Mott
Christine Brisson
Alexandra Kinyon
Colin Warner

Corrected translations:
Susan Converse
Tsan-Kuang Lee

-----------------------------
Ann Bies 
bies@ldc.upenn.edu
Linguistic Data Consortium
January 9, 2007
-----------------------------