Syllabus

1. Introduction
2. Characteristics of natural language
3. Tools and techniques
4. Applications

Rationale for Course Design
Natural Language Processing has progressed from working with simple rules on a few sets of sentences to being an important component of the search, retrieval, summarization, and presentation activities you use every day. I would like you to learn more about the established tools and techniques used behind the scenes, as well as the (many) remaining hard problems and the applications we may see in the future.

The first part of the course will focus on various language components and the tools and techniques that address them. We will use a variety of tools for in-class and take-home exercises. During the second part of the course, we will discuss the broader applications that use these tools and techniques.

Finally, you will explore a topic of your own interest in a project or paper at the end of the semester. This may build on an idea, application, tool, or technique we discussed earlier, or may focus on something we didn't have time to address.

I have been fascinated with words, language, and what people do with language for many years. Applications of technology to language is equally enthralling. I look forward to the opportunity to share my interests with you!

Course Objectives

1. Gain an appreciation of the wonderful complexity of natural language.
2. Gain hands-on experience with basic NLP tools and techniques.
3. Survey major applications of natural language processing. Important aspects of each application include the goals of the task, the materials and methods used, the state of the art, and how to evaluate progress in the field.
4. Explore resources for natural language processing.
5. Provide an opportunity for individual investigations in chosen topics.

1. Participation
   • I will always be prepared for class, and will start class on time. If unforeseeable circumstances prevent this for any reason, I will try to notify you beforehand if at all possible.
   • I expect the same of you: be prepared for class, be ready to start class on time, and let me know if you know you won't be there. If you must miss a class session unexpectedly, get in contact with me ASAP. While I do not necessarily take attendance, I pay attention to attendance.
   • You are encouraged and expected to be an active participant in class, including full class discussions, small group discussions and exercises, and the class list. Similarly, if you choose to do a group project, you are expected to be an active participant in your project team.
   • Reading assignments should be done before the class for which they are assigned so you can ask questions and participate in discussions.
   • If there is something you don't understand ask a question! If you don't want to ask during class, post it to the class list, come to my office hours, or contact me by email.
   • If you find an interesting or useful tool, collection, demo, website, article, etc., please send me a link or reference so I can add it to the Resources page.
   • Participation counts for a percentage of your final grade, but there are many ways to participate. Quality of contributions is more important than sheer quantity.
2. Assignments
   • Assignments must be submitted at the beginning of the class in which they are due.
   • A late assignment will be penalized 5% for every day it is late. If you have a real problem submitting an assignment on time, please talk to me before the due date.
   • The assignment descriptions will say how each assignment should be submitted (e.g., on paper, as an email attachment, as a webpage, etc.).
   • Pay attention to detail! Proofreading and clarity of presentation are important for information professionals.
   • Start working on assignments well in advance of the due date. It is usually a good idea to give yourself time to let your work sit for a bit, then come back to review it. I take a dim view of questions that show you are starting an assignment only 24 hours before it is due. Besides, what if I were unavailable for consultation?
3. Giving and receiving help.
   • If you are having difficulty with an assignment or some aspect of your project, please come and see me!
   • The Honor Code, which prohibits giving or receiving unauthorized aid in the completion of assignments, is in effect in this class. The Instrument of Student Judicial Governance gives examples of actions that constitute academic dishonesty. There are some specific guidelines for this class.
     • You may give and receive assistance regarding the use of hardware and software.
     • I encourage you to discuss issues raised in class or by the readings with each other. You may also ask your classmates for clarification of class notes.
     • Individual assignments are to be done individually. You may consult the course readings, your notes, and even other print or web sources. (Keep in mind, however, that what you find in other sources may not be consistent with what I want you to do.) You may not consult your classmates or other people; all questions should be addressed to me.
     • The exercises and final project may be done in pairs, with both members taking responsibility for the products. Work should be distributed equitably between each member.
4. Electronic devices in class.
   • You do not need to bring your laptop to every class. I will remind you in advance for the classes for which you do need your laptop.
   • You will need your laptop for the classes in which we start exercises, and you may find them helpful for playing with demos as well. I will remind you when you need to bring them.
   • You will need to load the GATE tool set on your laptop for the exercises and GATE assignment. You may wish to load additional tools as well -- not all tools support web access. (This is one of the things you can help each other with.)
   • If you plan to use your laptop to take notes, be aware that I sometimes sketch diagrams on the board. You may want to bring paper and a writing utensil so you can copy them quickly.
   • Laptops, PDAs, and similar devices should be used only for class purposes. Unrelated activities such as reading email, IMing, surfing the web, or playing games are distracting and discourteous.
   • Please remember to turn off your cellphone before class starts.

I will send out a test message or two the first week of class. I will use the list to send out announcements. You can use the list to ask questions of the class in general, to share helpful hints about software, etc. To post to the list, you can send email to inls512_001@listerv.unc.edu, or go to the list's web page.

Exercises 1-5, 5% each, 25% total
GATE assignment, 25%
Project, 40%
Participation, 10%

All assignments will be graded on the following scale (graduate/undergraduate):

• 95 - 100% H / A
• 90 - 94% P+ / A-
• 85 - 89% P / B+, B, B-
• 80 - 84% P- / C+, C
• 70 - 79% L / C-, D
• 69% and below F