Announcements | Course Information | Schedule
[画像:parse trees]
http://cs.jhu.edu/~jason/licl).
This class presents fundamental methods of computational linguistics. We will develop probabilistic models to describe what trees and sequences are likely in a language. After estimating the parameters of such models, it is possible to recover underlying structure from surface observations. We will examine algorithms to accomplish these tasks.
We will also survey a range of current tasks in applied natural language processing. Many of these tasks can be addressed with techniques from the class. Some previous exposure to probability and programming may be helpful. However, probabilistic modeling techniques will be carefully introduced, and programming expertise will not be required. We will use a very high-level language (Dyna) to describe algorithms and visualize their execution.
Useful related courses include Machine Learning, Python 3 for Linguists, Corpus-based Linguistic Research, and Computational Psycholinguistics.
[1] Or are these statistics? What are "statistics" anyway? How are they different from "information"?
Note: Lecture schedule below is tentative. Assignments and readings will be added.
Note: Additional slides plus more computationally intensive assignments are available from my NLP course at Johns Hopkins, which is more than twice as long and assumes more computational background (but no prior linguistic background). Here's an old list of other NLP courses.
Last Modification $Date: 2013年09月02日 06:43:30 $ (GMT)