The main readings will be:

- Speech and Language processing by Jurafsky and Martin. (main text)
- Foundations of statistical natural language processing by Manning and Schutze. (Suggested if you have a weaker statistics background.)
- Language log (In particular, posts by Mark Liberman often have a lot of statistical content.)

- Homework 1 is due Sept 24th.

- Sept 5: Introduction (.pdf)

- Sept 10: Regular expressions (.pdf)
- Sept 12: Ngrams (.pdf)
- N-grams (Chapter 4 of JM)

- Sept 13 at noon: Justin Rising and Josh Magarick are running a session called "Python for Statisticians". Lunch will be served!

- Sept 19: High diminsional reasoning
- Read before class:

- Sept 21: Lyle on eigenwords (in the old room SHDH 105)

- Sept 24: Speech: encoding
(.pdf) (chapters 7 and 8)
- NEW ROOM! R2D2 441
- Limits on D to A: Shannon limit
- Limits on A to D: Nyquist rate
- IPA (fancy alphabet: read along as various people speak words.)
- Homework 1 is due
- the regular expression as a single file

- Sept 26: Speech: decoding (chapters 9 and 10)

- Oct 1: Backoff and information theory.
- Oct 3:

- Oct 8: Jordan: guest lecture on HMM's
- Homework 2 due date

- Oct 10: Adi on variable length markov chains

- Oct 15: Context Free Grammars
- Oct 17:

- Oct 22: Homework 3 due
- Oct 24: Risk inflation

- Oct 29: No class: rain day
- Oct 31: Streaming methods
- lecture notes
- Papers
- A good introduction to the ideas of risk ratios (Dongyu Lin)
- Bankruptcy example (67000 regressors)
- mFDR (with martingale proof)
- the martingale

- Nov 5: The power of large blocks
- risk graphic (finishing last weeks class)
- slides for Dongyu's job talk

- Nov 7: Parsing

- Nov 12:Statistical parsing
- Nov 14: notes

- Nov 26: Machine traslation (chapter 25)
- Read chapters 17, 18 and 25
- nice tutorial on IBM models 1,2,3, by Kevin Knight

- Nov 28: CCA
- slides for today's lecture
- paper with Sham
- CCA goes back to the 1930's, so there should be pleanty of web material to look over. I won't put it up. But if you find something nice, email it to me and I'll post it.

- Dec 3: Disambiguation
- Other disambiguation solutions
- Read chapter 20

- Dec 5: Hadamard transformations?

dean.foster@gmail.com Last modified: Wed Dec 5 14:23:59 EST 2012