Instructor: Robert C. Berwick	TA: Karen Kohl
Email: berwick@ai.mit.edu	Email: ktkohl@ai.mit.edu
Office: 35-423	Office: NE43-809
Phone: (617) 253-8918	Phone: (617) 253-0778
Office hours: Weds 1-3 and by appointment	Office hours: Tu 5-7pm in 14-0637

IMPORTANT CLASS MESSAGES - New on this course's web pages:

• Lecture 22 (Language Acquisition Part 2) posted here and in 2-page bw format here.
  
• Lecture 21 (Language Acquisition Part 1) posted here and in 2-page bw format here.
• Lecture 20 (Machine translation 4) posted here and in 2-page bw format here.
  
• Lecture 19 (Machine translation 3) posted here and in 2-page bw format here.
  
• Lecture 17 (Machine translation 1) posted here and in 2-page bw format here. This has a  brief description of the word-word translation task below.  Lecture 18 (Machine translation 2) posted here and in 2-page format here.
• Lectures 16, 15, 14, 13, and 12 are posted in both full and 2-page bw formats:
• Lecture 12 & 2p/bw; Lecture13 & 2p/bw; Lecture 14 & 2p/bw; Lecture 15 & 2p/bw; Lecture 16 & 2p/bw
  
• Please do the word-word translation started in class Weds. Your job: translate sentence 13(b) into a corresponding, translated sentence in language A.  All the data you need are located here, a list of 13 paired language A and language B sentences, then a bigram table for additional language A sentences.  Please don't use the web - you don't need to, and it will spoil the fun.  If you are more ambitious, try translating  the additional two sentences 14(b) and 15(b) into their language A counterparts. Come to class with your results, if you can; otherwise, send a URL to Karen.  If you send a URL, you'll have to include some description of how you arrived at your results.
  

Prerequisites & Relation to Other Courses

Assignments, Due Dates, and Policies - General Description




Assignment Schedule:





No.	DUE DATE	Task	Resources
#1	Feb 26	Word Parsing	Laboratory 1 material is here.
#2	Mar 05	Word Tagging	Brill part of speech tagger; web version of Brill tagger here. Paper on Brill tagger here. Hmm tagger paper here.
#3	Mar 21	Phrase Parsing	deMarcken parser; documentation here
#4	Apr 09	Semantic Interpretation	Syntax-directed semantic interpretation system; laboratory here.
#5	Apr 09	Probabilistic methods; IR	Probabilistic parsers and extensions of HMM, other association methods CANCELLED
#6	Apr 23	Enrichment  Report	Your own choice, an 'engineering evaluation' reporting on an extension to any of the labs/topics in the course  - CANCELLED
#7	May 14	Final Project	Your own design with team partners




Open to submissions

Closed to submissions (you can still hand in, but subject to 'late days'




Assignment Specifics:


Laboratory 1 - Word parsing & morphology

        Laboratory 2 - Word tagging & comparing statistical models

        Laboratory 3 - Phrase structure parsing

        Laboratory 4 - Semantic interpretation

       Tentative Course Schedule:

Week 1	INTRODUCTION: The NLP enterprise, from words to meaning	Reading in Textbook or Notes
W 02/05	Introduction, Organization, Laboratories. Course Overview: Intro to NLP. Main Issues; fsa's [pdf]  [4slides/page for printing] For fun, try this link: postmodern. Reference for this generation program is  here.	pp.1-57;  Notes 1
Week 2	WORD MODELING: automata and linguistics	pp. 58-90
M 02/10	Linguistics: Phonology and Morphology I.; 2-level morphology, Kimmo  [pdf]  [2slides/page for printing]  Lab 1 intro: processing words	Notes 2
W 02/12	Linguistics: Phonology and Morphology II.  [pdf]  [2slides/page for printing]  Lab 1b : processing  Spanish words	pp. 287-321
Week 3	WORD MODELING: statistical approaches & part of speech tagging
T 02/18	President's Day Class (MIT turns Tuesday into Monday) OOPS SNOW DAY!!	pp. 235-284
W 02/19	HMM Tagging; Statistical Transformation Rule-Based Tagging; Precision, Recall, Accuracy For fun (and learning) try this link to an online Brill tagger here. Note: Default tagging is for Swedish! Click 'English' and click tracing 'on' if you want to see how it works.
Week 4	LINGUISTICS & GRAMMARS; PARSING ALGORTHMS I	Notes 3
M 02/24	Part of Speech Tagging: the Brill Tagger. Lab 1 due; Lab 2 handed out.	pp. 357-394
W 02/26	Introduction to Parsing; Linguistics: Syntax & Parsing .
Week 5	PARSING ALGORITHMS II	Notes 4
M 03/03	Shift-Reduce Parsers in Detail. Earley's Algorithm and Chart Parsing
W 03/05	Context-free Parsing and Beyond: Efficiency Issues; Feature-based parsing; NL system design Lab 2 due.	pp. 477-498
F 03/07	Add Date
Week 6	PARSING ALGORITHMS and the LEXICON
M 03/10	Shift-Reduce Parsers in Detail. Earley's Algorithm and Chart Parsing  Lab 3 handed out
W 03/12	Parsing with an integrated lexicon - the question of syntactic features.
Week 7	SEMANTIC INTERPRETATION	pp. 395-446;pp. 447-476
M 03/17	Semantic Interpretation I: compositionality
W 03/19	Semantic Interpretation II: compositionality and quantifiers
3/24-3/28	No classes - Spring Break. Go to someplace warm if you can.Lab 3 due March 21, Lab 4 handed out March 22
Week 8	FEATURE Parsing; TREE BANKS & PROBABILISTIC PARSING	Notes 5
M 03/31	Semantics III: Lexical Semantics
W 04/02	Semantics IV: Lexical Semantics.	pp. 501-544
Week 9	SEMANTICS II
M 04/07	Semantics V: Constraint-based systems	pp. 545-588
W 04/09	Lab 4 due
Week 10	MACHINE TRANSLATION I & II	pp. 589-630
M 04/14	Machine Translation I
W 04/16	Machine Translation II
Week 11	MACHINE TRANSLATION III
M 04/21	No class - Patriot's Day.  Run the Boston Marathon if you can.
W 04/23	Machine Translation III.   "Enrichment" Lab  due (on or before) - CANCELLED	pp. 631-666
Th 04/24	Drop Date
Week 12	MACHINE TRANSLATION, IV	Notes 6
M 04/28	Project Discussion Day
W 04/30	Machine Translation IV	pp. 799-830
Week 13	LANGUAGE LEARNING
M 05/05	Language learning I
W 05/07	Language learning II
Week 14	EVOLUTIONARY MODELS OF LANGUAGE LEARNING & ORIGINS
M 05/12	Computational Models of Language Change, I
W 05/14	Final project due.  Computational Models of Language Change and the Origins of Language

---

Grading Weights:




Assignments	55%
Final Project	35%
Class Participation	10%