next up previous
Next: Reading material Up: Elements of Language Processing Previous: General information

Description of course

Elements of Language Processing and Learning
This course will be given in English!

Lecturers: Khalil Sima'an

Summary The course covers the basics of structured (graphical) statistical computational linguistics models for sentence parsing, language modeling and machine translation. The course concentrates on statistical modeling of two kinds of data: labeled, ordered trees and parallel translation corpora. From the statistical point of view the course looks at models for which the training data is complete (PCFGs from treebanks) and also models for which the data is incomplete (hidden PCFG derivation models, DOP for treebanks; and parallel corpora for translation models with word alignment and tree-level alignment).



Khalil Sima'an 2012-10-29