In the past, the makeup of students taking this course has been distributed between linguists and other students of the humanities who may have had little or no experience with computers, and computer scientists who may have had little or no experience in linguistics. In order to accomodate the whole range of students in this class, we will necessarily limit the depth to which we pursue both the computational and linguistic disciplines.
There will be no programming required for the course, nor will the students be required to read code, but the emphasis will be on the data structures which underlie the programs and the code. This course will not dwell on many mathematical formulas, and those that we do address will be firmly grounded in examples. Where mathmatics is involved, the emphasis will be on understanding how and why a given formula expresses what it does, rather than making calculations (that's what the computers are for). Students' interactions with computers will consist of using a word processor, making use of the internet, and using the computer tools in the multimedia lab.
The linguistic theories we deal with will also necessarily be visited on a bit of a whirlwind tour in the hope that students will come back for a closer look at the areas that have a special appeal. There will be ample references to more technical material for those who want to deal with some of the domains introduced here in more depth.
The text we'll be using for the course is Natural Language and Computationl Linguistics, by Colin Beardon et.al.Those who are interested in pursuing this subject in greater technical depth might want to start with Survey of the State of the Art in Human Language Technology available as postscript, This book covers a very broad range of topics, each chapter starts with a relatively accessible introduction, and descends gradually deeper into more technical elaborations. At the end of each chapter is an excellent bibliography of the subject.
There will also be other suplementary material put on reserve in the library.
The multimedia lab has two wings: one in BA-419 (phone:594-5370), and the other in BA-410 (phone:594-8010). Both labs will be open from 8am to 4:30pm, and M-Th, one of the labs will be open from 8am to 8pm, but which wing stays open late will vary from day to day. Also, there are many occasions where the lab is reserved, so you should call ahead if you're planning on using it. It is equiped with computers (Windows95 and MacOS), word processors, and internet access,as well a wealth of CD-ROMS dealing largely with language learning.
In doing your research for your papers, a potentially good place to go for references on your topic is USENET newsgroups. You can subscribe to these with Netscape or Microsoft Outlook. If you post to a newsgroup, tell them the research you're doing and ask for references, there's a good chance that someone in the know will be able to give you some leads, especially if you demonstrate that you've already looked around a bit yourself. one newsgroup you might find particularly helpful is comp.ai.nat-lang.
[ Up to Tests, Assignments, and Grading]
There will be two mid-terms, each worth 20% of your grade. As each lecture is given, the course outline at the website will contain links to an outline of the lecture with a corresponding set of quiz questions. Midterms will be strongly derivative of these quiz questions.
[ Up to Tests, Assignments, and Grading]
There will be occasional assignments involving interaction with computer tools. Among other things, will be playing with a voice analyzer, and with a lexical database known as WordNet These will amount to another 20% of your grade.
[ Up to Tests, Assignments, and Grading]
There will be a term paper due at the end of the term. Students are free to choose the topic (as long as it is pertinent to this course), and encouraged to keep it as close to their own interests as possible (e.g. a business major may want to write a sketch of a business plan dealing with some NLP application). Minimum 10 pages. Unless this will cause the student some undue hardship, the paper should be submitted in HTML. The last week of the course and final will be set aside for each student to discuss his or her paper. Exemplary papers will be invited to be posted on the course web site at the end of the term.
Note that there will be no final, we will use that time for presenation of papers.
