CS 8761 -Natural Language Processing - Fall 2002

Instructor: Dr. Ted Pedersen
Office: 309 Heller Hall
Office Hours: Mon & Wed 4:45pm-6:00pm, Thu 2:00pm-3:00pm
Email: tpederse@umn.edu

Course Web Page: http://www.d.umn.edu/~tpederse/Courses/CS8761/class.html

Course Objectives:

Natural Language Processing seeks to analyze, generate, and understand human language via computational techniques. This course focuses on empirical approaches to lexical and syntactic analysis, semantic interpretation, and discourse processing. Specific applications include part-of-speech tagging, machine translation, and authorship attribution.

Required Text:

Foundations of Statistical Natural Language Processing by Christopher Manning and Hinrich Schutze. MIT Press. There is a supporting Web Site with quite a bit of information.

There is a copy of the text on 2-hour reserve in the library. You will still need to have your own copy of the text, however, this might prove useful if you forget your book, etc.

Reading assignments will be given in the lecture and posted here.

Suggested Texts:

We will do our programming assignments in Perl. While we will discuss Perl from time to time in the lecture, there will be a fair bit of self-study required. As such you are strongly advised to have at least one of the following at your disposal:

Learning Perl (3rd Edition) by Randal Schwartz and Tom Phoenix. O'Reilly Publishers. You can get this book from amazon.com or most any bookstore. This takes a tutorial approach and is especially good if you have limited C or Unix experience. The 2nd Edition is available in our library as an electronic book. The local Barnes & Noble has several copies of the 2nd edition in their bargain area for $7.98 (as of Aug 27, 2002).

Programming Perl (2nd Edition) by Larry Wall, Tom Christiansen, Randal Schwartz and Stephen Potter. O'Reilly Publishers. You can get this book from amazon.com or most any bookstore. This book is more like a reference manual than Learning Perl, although it is still very readable. This is a good choice if you have extensive C and Unix backgrounds. There is a copy of this book on two hour reserve in the library.

Mastering Regular Expressions: Powerful Techniques for Perl and Other Tools (2nd Edition) by Jeffrey E. Friedl. O'Reilly Publishers. You can get this book from amazon.com , or any other bookstore. This is an in-depth treatment of regular expressions in Perl and is the best available reference on this topic. You can get an electronic copy of the 1st Edition of this book from the UMD library.

There are two other textbooks on 2-hour reserve in the library. The first is by Daniel Jurafsky and James Martin, Speech and Language Processing. The second is by Eugene Charniak, Statistical Language Learning. Both are excellent textbooks. The Charniak book focuses on empirical methods and is an excellent complement to our required text. The Jurafsky and Martin book is more general in nature and includes some discussion of speech processing.

Prerequisites:

This class is only open to currently enrolled first-year CS graduate students.

Grading Basis: Grading Scale: Programming Assignments:

Programming assignments are to be completed in Perl and must be submitted on time. Late work is not accepted and will result in a score of zero for that assignment. You must use the web drop link on the class web page to turn in assignments.

Each programming assignment is worth 10 points. There will be 3-5 programming assignments.

All programming assignments are individual. You are expected to write your own code. If you turn in code that is not your own (e.g., code taken from a book or online archive, code written by a colleague or classmate, etc) I reserve the right to give you a failing grade for the semester.

Final Project:

You will be assigned to a team and given a challenging problem in natural language processing to tackle. You must deliver a software solution and a final report that includes a discussion of your team's solution, an evaluation of its effectiveness, and a survey of related work. All teams will work on the same problem and we will have a comparative evaluation to see how well each team fares relative to the others.

You are expected to collaborate and work closely with your teammates. You may not collaborate with anyone outside of your team for any reason. All members of a team will receive the same grade. Your team must produce original work. Any type of plagiarism, whether it is deliberate or accidental, will be dealt with harshly.

Exams:

All exams are closed-note, closed-book. You must take exams at the scheduled time and place. Exams will not be given early. Make-up exams will only be offered in the event of documented personal emergencies. The final exam must be taken at the date and time as determined by the official university schedule (Monday, December 16).

Quizzes

All quizzes are closed-note, closed-book. They may cover any topic discussed in the lecture or included in the assigned readings. They will not be announced ahead of time. If you are not in the lecture at the time the quiz is given you will receive a 0. Your low quiz score will be dropped. We will have at least 8 but no more than 12 quizzes. Each quiz will be worth 10 points.

Lecture Notes:

After dutifully posting lecture notes for a few classes, I have concluded that it isn't especially helpful, and that it might actually be a bad thing. In some cases the availability of online lecture notes clearly acts as a deterrent to attending lecture and/or taking careful notes. Since these are important skills that you should develop and hone while in school, I won't be posting or distributing my lecture notes.

However, if you have some temporary or permanent disability that affects your ability to take notes then please let me know and I will make alternate arrangements with you.

Equal Access:

If you have any disability (either permanent or temporary) that might affect your ability to perform in this class, please inform me at the start of the semester or as soon as you learn of such a condition. I may adapt methods, materials, or testing so that you can participate equitably. To learn about the services that UMD provides to students with disabilities, contact the Access Center, 138 Kplz, phone 8217 or visit their web page.

By: Ted Pedersen - tpederse@umn.edu