|
Courses and Workshops
|
| Title: | Statistical methods for natural language processing |
| Lecturer(s): | Helmut Schmid (University of Stuttgart) |
| Type: | Foundational Course |
| Section: | Language and Computation |
| Week: | First | | Time: | 14.00-15.30 (Slot 3) |
| Webpage: | http://www.ims.uni-stuttgart.de/~schmid
|
| Reader: | Course will not have a reader | | Room: | A3 |
Description This course is intended to provide an introduction to basic methods in
statistical NLP like N-gram models, Hidden Markov Models and
(Lexicalised) Probabilistic Context-Free Grammars and their efficient
computational treatment with dynamic programming algorithms (Viterbi,
Forward-Backward and Inside-Outside algorithm).
It presents applications of these methods for language modelling,
language guessing, part-of-speech tagging, and probabilistic parsing and
also discusses implementational issues like parameter estimation and
smoothing.
If there is sufficient time, I would like to add one or two more
advanced topics like generative probability models for unification-based
grammars or special HMMs for phrase-break prediction and syllabification.
The course is intended for students with basic knowledge of formal
language theory and statistics.
|
© 2003, 2004 LORIA, Carlos Areces. |
2003-11-20
| |
