CSC 401, Winter 2008

Natural Language Computing

Resources for Assignment 1

Check the 401 home page and the course bulletin board regularly for course announcements, assignment hints, etc.

• The Brown Corpus (gzipped tar file for Unix / Linux, 2.9Mb)
• The Brown Corpus (zipped file for PC, 3.2Mb)
• The Brill tagger (gzipped tar file for Unix / Linux, 1.3Mb)
• Brill tagger for Windows -  you have a few options here:
• get cygwin and install the Brill tagger for cygwin
• install the Python dll module for the Brill tagger
• stop using Windows :-)
• Scripts:
• Readme file for scripts
• RunBrown2Brill
• RunTagger
• RunGatherStats
• splitre (used by RunTagger)
• Word lists:
• English abbreviations
• First person
• Second person
• Third person
• Conjunct words
• Verbs of saying
• Private verbs
• Public verbs
• C4.5 package, release 8 (approx 145Kb). The link is at the bottom of Professor Quinlan's home page; you get a gzipped tar file with source code and installation instructions. Note restrictions on distribution! This package runs under Unix / Linux only; not for Windows.
• BrownStats.cases

Python

Other resources

• Oxford English Dictionary (2nd edition), on-line search (from UofT IP addresses only).