MIT Electrical Engineering and Computer Science Fall 2003 Catalogue Supplement

6.891 Machine Learning Approaches for Natural Language Processing (H)

L MW 4-5:30, Room 1-379
Prof. Michael Collins, Room NE43-723, 3-0705
Prereq.: 6.034 or permission of instructor
3-0-9

This subject qualifies as an Artificial Intelligence and Applications Concentration subject

A survey of statistical or machine learning approaches to natural language processing, with an emphasis on research issues. Techniques covered will include hidden markov models, stochastic context-free grammars, boosting and maximum-entropy methods, kernel methods over natural language structures, and learning from unlabeled or partially labeled data. Problems covered will include statistical parsing and tagging problems, information extraction, machine translation, dialogue systems, and other NLP problems. 3 Engineering Design Points.