|
|
Ling 236: Quantitative,
Probabilistic, and Optimization-Based Explanation in Linguistics |
|
Course Information |
Lecture: 1-4 units, W 2:15-5:05 Art 4 . [My understanding is that this is a funky room in the basement of the Cummings Art building. We'll see how it works out.]
Useful Information and Handouts
Course Description
This course will be a combined tutorial and seminar on the soft constraints inherent in linguistic systems. It will attempt to justify that there are such constraints, and to examine how they can be modeled, based on quantitative evidence obtained from linguistic corpora or experiments. The course will focus on syntax, and the aim is to build models that achieve linguistic explanation. (Contrary to the catalog description) I’ve decided not to cover collecting data. It is assumed that you have or can get some raw data to work with. If you think you need more skills in collecting data, you should consider another course that covers this, such as Ling 121, Ling 237, Ling 258, Ling 286, Psych 214, etc. Some of these are on this quarter. However, the tutorial part is that I do intend to present and teach methods for statistical data analysis and for building probability models of language. Since this is the kind of material that I think people don’t learn without using it, there will be small weekly assignments. The seminar part will be reading and discussion of recent and older papers that develop quantitative approaches to language, and the presentation and development of ideas for probabilistic models of language.
Course Objective
For students to be able to understand and build probabilistic models of linguistic phenomena.
Contact Info
Christopher
Manning
Office: Gates Bldg. Room 418
Office Hours: TBA
Phone: (650) 723-7683
Fax: (650) 725-2588
E-mail: manning@cs.stanford.edu
Prerequisites
- A course in syntax
- Some quantitative data, or the knowledge and means for collecting some
Intended Audience
Graduate students and advanced undergraduates specializing in linguistics or symbolic systems.
|
Reading and Work |
Pre-reading
If you’d like to get a head start on things it’d be good to know about in this course, here are the recommendations:
- If you know little about
probability, then it’d be good to know more! One good place to start would
be John Goldsmith’s tutorial, Probabiity for linguists available in Microsoft
Word or converted
to HTML. Much the same
material and some additional material is covered more concisely in chapter
2 of Christopher Manning and Hinrich Schütze,
Foundations of Statistical Natural Language Processing. MIT Press, 1999. You can read the first few pages at Amazon. - If you’d like to get into the kind of material that will be dealt with in this course, here’s a draft survey chapter on Probabilistic Syntax by me, available in PDF or PS format.
Reading
There is no required text. Readings and handouts will be distributed. Copies of materials available electronically will be posted here, and hard copies will be distributed in class.
Work and Grading
The course work will be:
- 8 small assignments [written, may require use of computers/Web] (4% each)
- 1 final paper prospectus (8%)
- 1 final paper on a topic of your choice within the domain of the course (60%)
The other expectation is participation in class presentations and discussion.