Information Retrieval meets Statistical Language Technology
Lecture: T Th, 12:00-1:20pm, EE1 026 (campus maps may be found here.)
Tues., January 12 - Handout: http://www.infotoday.com/searcher/jun/story4.htm
Thurs., January 14 - No Class
Tues., January 19 - Chapter 5: Collocations
Thurs., January 21 - Chapter 7: Word Sense Disambiguation
Tues., January 26 - Chapter 7: continued.
Thurs., January 28 - No Class
Tues., Feb 2 - Chapter 8: Lexical Acquisition
Th., Feb 4 - Roark + Hearst papers
Tues., Feb 9 - Class cancelled [presenter ill]
Th., Feb 11 - MindNet
Tues., Feb. 16 - Vorhees paper + overview of Wordnet.
Thurs., Feb. 18 - Industrial affiliates. No class
Tues., Feb. 23 - Latent Semantic Indexing (ch. 15).
Thurs., Feb. 25 - No class.
Tues., March 2 - Text Categorization (ch. 16).
Thurs., March 4 - Clustering (ch. 14).
Tues., March 9 -- Information Extraction.
| Office Hours | Email Address | |
|---|---|---|
| Oren Etzioni, Instructor | By appointment, Sieg 209 | etzioni@cs.washington.edu |
| Patrick Allen, CSE Staff | Sieg 213 | pjallen@cs.washington.edu |
Christopher D. Manning and Hinrich Schutze, Foundations of Statistical Natural Language Processing, MIT Press, Draft of January 5, 1999. Website: http://www.sultry.arts.usyd.edu.au/fsnlp
See Erik for the TREC CDs.
TREC Website: http://trec.nist.gov
Parser + Adam's notes: http://www.cs.washington.edu/homes/carlson/courses/cs590q-wi99/mindnet.html
Taggers, corpora, more: http://www.sultry.arts.usyd.edu.au/links/statnlp.html
C&C maintains a number of statistical software packages on their computers: http://www.washington.edu/computing/software/profiles/sw_stat.html
Some of these packages, including S-Plus, generally considered the most fully-featured of the bunch, are also available in the CS&E departmental labs in Sieg. Also, Excel can compute a number of statistical measures.
A main page explaining how to use the MetaCrawler engine, with all the neat command switches you can use and exploit, is available at: http://huskysearch.cs.washington.edu/dev/man/metacrawler.html
In order to use the MetaCrawler as a binary, you need to:
1) Send mail to whj and get an AFS account.
2) Send mail to me when whj has set you up with one.
You'll be given an account on the MetaCrawler machines (vorlon, draz, zhadum, centauri-prime, and minbar) and will be able to use it there.
Marc Friedman Lexical Acquisition
Adam Carlson Latent Semantic Indexing or How I Learned to Stop Worrying and Love Math I Don’t Understand
Geoff Hulten Chapter 3 : Corpus-Based Work