Natural Language Processing (CS-5340/6340)
Fall Semester 2006 - REVISED SCHEDULE

Lectures:	Tuesday and Thursday, 2:00-3:20, EMCB 120
Textbook:	Speech and Language Processing: An Introduction to Natural Language
	Processing, Computational Linguistics, and Speech Recognition
	by Daniel Jurafsky and James Martin
Instructor:	Professor Ellen Riloff
E-mail:	teach-cs5340@cs.utah.edu (preferred) or riloff@cs.utah.edu
Office:	3140 MEB
Consulting Hours:	anytime when my door is open, and by appointment
TAs:	Bryce Anderson, Sean Igo
E-mail:	teach-cs5340@cs.utah.edu (preferred)
	or andersbr@cs.utah,edu, sigo@cs.utah.edu
Consulting Hours:	by appointment

Instructor and TAs mailing list:

teach-cs5340@cs.utah.edu
Mail to this list goes to the instructor and the TAs. Please send questions to this list so that we can keep track of the questions from and responses to the class, and so that you will get the quickest possible answer!

Class mailing list:

cs5340@cs.utah.edu
Please subscribe to the class mailing list as soon as possible by signing up on the following web page: http://mailman.cs.utah.edu/mailman/listinfo/cs5340

Course web page:

http://www.cs.utah.edu/classes/cs5340
Our course web page contains pointers to the syllabus, lecture slides, assignments, solution sets, and general resources for more information about natural language processing.

Note to cs6340 students:

all mailing lists and course info will be indexed under the cs5340 name.

Course Goals

The goals for this course are to study:

• algorithms and methods for building computational models of natural language understanding, including parsing techniques, semantic representations, memory models, discourse processing, and statistical and corpus-based methods for automated knowledge acquisition.
• issues involved in natural language understanding, such as cognitive and linguistic phenomena.
• applications that can benefit from natural language processing, such as information extraction, question answering, machine translation, and spoken language understanding.

By the end of the course, we hope that everyone will have a good understanding of and appreciation for natural language processing, and have the necessary skills to build natural language processing tools.

SCHEDULE (subject to change)
All readings are from the textbook unless otherwise specified.

Date	Topic	Reading
AUG 24	Introduction to and history of NLP	Ch. 1, 2
AUG 29	Syntax: the basics	Ch. 3.1-3.2 (pp. 57-70), 3.4-3.6
		Ch. 8.1-8.4
AUG 31	Syntax: chart parsing	Ch. 10.4
SEP 5	Syntax: transition network parsing	Ch. 9.1-9.8, 9.10-9.13
		Ch. 10.1-10.3, 10.5-10.6
SEP 7	Syntax: parsing cont'd
SEP 12	Probability; N-gram models	Ch. 6.1-6.3 (pp. 191-210), 8.5
SEP 14	Probabilistic algorithms	Ch. 5.9 (pp. 169-179)
SEP 19	Probabilistic algorithms (cont'd)
SEP 21	Probabilistic context-free grammars	Ch. 12.1-12.3
SEP 26	Experimental design	pp. 315
SEP 28	Information extraction	Ch. 15.5
OCT 3	Semantic lexicon bootstrapping
OCT 5	FALL BREAK
OCT 10	Semantics: the basics	Ch. 14, 16
OCT 12	Semantics: the basic (cont'd)
OCT 17	MIDTERM EXAM
OCT 19	Learning extraction patterns	Handout
OCT 24	Semantics: named entity recognition
OCT 26	Discourse	Ch. 18.1
OCT 31	Semantics: word sense disambiguation	Ch. 17.1-17.2
NOV 2	Transformation-based learning	Ch. 8.6
NOV 7	Conceptual dependency theory	Handout
NOV 9	Applications: information retrieval and Q/A	Ch. 17.3-17.5
NOV 14	Applications: speech understanding	Ch. 7.1-7.3, 9.9
NOV 16	Applications: machine translation	Ch. 21
NOV 21	Summarization (guest lecture by $Hal Daum\acute{e} III$)
NOV 23	THANKSGIVING BREAK
NOV 28	Project presentations
NOV 30	Project presentations
DEC 5	Project presentations
DEC 7	Project presentations
DEC 11	FINAL EXAM 1:00-3:00 in EMCB 120

GRADING POLICY

Grades for the course will be based on homework assignments (both written questions and programs), a project, a midterm exam, and a final exam. Overall grades will be based on the following formula:

Written Assignments:

16% of final grade.

Programming Assignments:

24% of final grade.

Exams:

30% of final grade.

Project:

30% of final grade.

Homework assignments must be submitted using the electronic handin system, including BOTH written and programming assignments. The written assignments must be submitted in one of the following formats: pdf, postscript, html (without hyperlinks), or plain ascii text. Instructions for using the electronic handin system will accompany each assignment. Assignments slipped under a door or left in the main office will not be accepted.

Programs may use any of the following programming languages: C++, Java, or python. You must get permission from the instructor if you wish to use any other programming language. All programs must compile and run on the linux-based CADE machines (lab1 and lab4 machines). You will receive ZERO credit for programs that do not compile and run on the linux-based CADE machines. This policy is necessary for the instructor and TAs to test and grade your programs ourselves.

LATE POLICY

Late assignments will not be accepted. The electronic submission system will be disabled promptly after the time specified as the due date for each assignment. No assignments will be accepted thereafter. Our motivation for this strict late policy is to ensure that assignments are graded and returned as quickly as possible. Allowing late assignments holds up the grading process and the distribution of solutions. We will do our best to grade the assignments and hand out solutions promptly so that everyone will get feedback in a timely fashion.

CHEATING POLICY

Cheating of any kind will not be tolerated. Any assignment or exam that is handed in must be your own work!! However, talking with one another to understand the material better is strongly encouraged. Recognizing the distinction between cheating and cooperation is very important.

• If you copy someone else's solution, you are cheating. If you let someone else copy your solution, you are cheating. We will not distinguish between the person who copied a solution and the person whose solution was copied. Both people will be treated as cheaters.

• If someone dictates a solution to you, you are cheating. Everything you hand in must be in your own words, and based on your own understanding of the solution.

• If someone helps you understand the problem during a high-level discussion, you are not cheating. We strongly encourage students to help one another understand the material presented in class, in the book, and general issues relevant to the assignments.

• When taking an exam, you must work independently. Any collaboration during an exam will be considered cheating.

• Any student who is caught cheating will be given an E in the course and referred to the University Student Behavior Committee. Many students think they can get away with cheating and will not be caught. But it is much easier to spot cheaters than you might think! Please don't take that chance - if you're having trouble understanding the material, please let us know and we will be more than happy to help.

AMERICANS WITH DISABILITIES ACT (ADA)

The University of Utah seeks to provide equal access to its programs, services and activities for people with disabilities. If you will need accommodations in the class, reasonable prior notice needs to be given to the Center for Disability Services, 162 Olpin Union Building, 581-5020 (V/TDD). CDS will work with you and the instructor to make arrangements for accommodations.

All written information in this course can be made available in alternative format with prior notification to the Center for Disability Services.