Goals

We will begin with a review of some classic MT systems, move on to the Noisy Channel model that has been so influential in statistical MT (SMT), and then cover basic components of a modern system, the word-alignment training, phrase alignment training, target language-modeling, and decoding. We will look at the contributions made by introducing classes and hierarchical synatctic information. Then we will read some papers and do some simple experiments with sense-disambiguation.

Course Outline

Here.

Required Text

The text for the class will be Jurafsky and Martin, Speech and Natural Language Processing, with some material from the 2nd Edition, focusing on Chapters 19 and 24, the word sense and MT chapters. There are also additional readings available online (see course outline).

Prerequisites and Grading

Prequisite: Some computer science or some linguistics; preferably Ling 581.

Grading will be based on exercises/projects a take-home midterm and final.

Assignments	30 %
Presentation	20 %
Final Project	50 %

Back to top.

Place and Time

Tu Th 11:00-12:15
SH-348
Storm Hall

Website

http://www-rohan.sdsu.edu/~gawron/mt_plus/mt

Contact Info

Mailing address:
Jean Mark Gawron
Department of Linguistics and Oriental Languages
San Diego State University
5500 Campanile Drive
San Diego, CA 92182-7727
Telephone: (619) 594-0252
Office Hours: Tu Th 16:00-17:30, BAM 321

Back to top.