• The Unicode Consortium: home page
• IPA transcription in Unicode
• The IPA in Unicode (data-entry tool)
• ``Automata via macros'' (Krishnamurthi)
• Stuttgart Finite State Transducer Tools
• Corpus of Contemporary American English
• SRI Language Modeling Toolkit
• ``An estimate of an upper bound for the entropy of English'' (Peter F. Brown et al.)
This courses examines computational techniques for producing and processing text in natural languages and introduces the theoretical basis for those techniques, both in linguistics and in computer science.
The class meets at 3:15 p.m. on Mondays, Wednesdays, and Fridays in Science 3819.
The textbooks for the course are Speech and language processing, by Daniel Jurafsky and James H. Martin (Upper Saddle River, New Jersey: Pearson - Prentice Hall; second edition, 2008; ISBN 978-0-13-605234-0), and A Semantic Web primer, by Grigoris Antoniou and Frank van Harmelen (Cambridge, Massachusetts: The MIT Press; second edition, 2008; ISBN 978-0-262-01242-3].
I'm John David Stone. My office is Science 3829. My office hours this semester are Mondays from 10 a.m. to noon, Wednesdays from 9 to 11 a.m., Thursdays from 2:15 to 4:15 p.m., and by appointment. I can be reached by telephone at extension 3181, or by e-mail to stone@cs.grinnell.edu.
Each student in the course is expected to prepare for and attend the sessions of the class, to read the assigned sections of the textbook carefully, to learn the concepts, distinctions, and methods that are presented there, to submit solutions to exercises requiring the application of those concepts, distinctions, and methods, and to take and pass a three-hour final examination over the subjects that the course deals with.
If you have specific physical, psychiatric, or learning disabilities and require accommodations, please let me know early in the semester so that we can work out appropriate arrangements.
The schedule of topics for the course includes reading assignments. Please read each specified section before the beginning of the class that follows it in the schedule.
From time to time, I'll propose exercises for you to solve. If the exercise involves writing a computer program, please submit your solution by e-mail. Construct a compressed tape-archive (.tgz) file containing your source-code files, any files containing documentation or non-public input data, and any files containing output from your test runs, and attach that file to a cover message.
For non-programming exercises, you may submit your answers either in hard copy or by e-mail. For e-mail solutions, I accept files in the following formats:
Note that I do not accept files in any of the .doc or .docx formats that various versions of Microsoft Word produce by default. If you're currently using Microsoft Word, I recommend that you download, install, and use the free OpenOffice.org software instead.
The final examination for this course will be held on Thursday, December 18, at 2 p.m. Please take this date into account when making your travel plans.
The examination will be an anthology of short problems and essay questions. You may use any books and any hard-copy notes that you bring with you to the examination session, but you may not use computers or other communications equipment.
I generally prepare a set of study questions to help students prepare for each class session. You may write up and submit answers to those questions for extra credit. If you miss a class session, you must write up and submit answers to the study questions for that session.
It is especially helpful if students raise for discussion any questions they may have about the day's topic, the assigned reading, or the exercises. I suggest that you write out such questions as part of your preparation for class sessions.
Your performance on the exercises during the semester will determine three-fifths of your grade, your performance on the mid-semester examination your performance on the final examination three-tenths, and class attendance and other imponderable factors one-tenth.
Given the choice between good work that is late and poor or incomplete work that is on time, I prefer the former. Therefore, if you submit a paper that is correct, but late, I generally do not impose a penalty. On the other hand, if you take extra time for an assignment and still turn in poor work, I judge it more harshly than if you had turned in the same paper on time.
There is, however, an absolute deadline: Once I return corrected papers for an exercise, or post a solution on the Web, I cannot give credit for subsequently submitted solutions to that exercise, though I will still examine and comment on them. Similarly, I cannot give credit for papers submitted after the solution has been discussed or presented in class, either by me or by a student, even if the submitter did not attend the class.
Since you will receive credit on the basis of your individual performance on the work assigned for this course, it would be unethical to submit the work of others as your own. You may, if you like, collaborate on solutions to exercises, but such solutions must be submitted jointly by all the members of the collaboration. You may not collaborate on the final examination or any part of it. If I encounter any indications of plagiarism, the Committee on Academic Standing will deal with them.
This document is available on the World Wide Web as
http://www.cs.grinnell.edu/~stone/courses/computational-linguistics/
This work is licensed under a Creative Commons Attribution-ShareAlike United States License.
created August 21, 2002
last revised August 18, 2008