Text: Jurafsky and Martin, Speech and Language Processing, 2nd edition.
| On | we will cover | which means that after class you will understand | if before class you have read |
|---|---|---|---|
| 9/1 | Introduction | programming for nlp | |
| 9/3 | Hidden Markov Models | viterbi, forward | JM 6 |
| 9/8 | Part of Speech Tagging | posterior decoding, forward-backward | JM 5; johnson |
| 9/10 | Expectation Maximization | L-KL=Q+H, Dirichlet Priors | |
| 9/15 | Variational Bayes, CRFs | digamma, f - E[f] | mccallum 2001 |
| 9/17 | CRFs, Signal Processing for Speech | log-sum-exp, fourier, MFCC | JM 9 |
| 9/22 | N-grams | Good-Turing, Katz, Kneser-Ney | JM 4 |
| 9/24 | Speech Recognition | forced alignment | JM 4 |
| 9/29 | Large Vocabulary Speech Recognition | beam search, rescoring | JM 10 |
| 10/1 | Context Free Grammars | highest prob string, infinite trees | JM 13 |
| 10/6 | Treebanks | right node raising | JM 12 |
| 10/8 | Parsing | cyk, earley | JM 14 |
| 10/13 | Parsing | posterior decoding, inside-outside | |
| 10/15 | Real-World Parsing | lexicalization, markovization, reranking | JM 15; charniak 05 |
| 10/20 | Discriminative Methods | tree CRF, max-margin parsing | taskar 04 |
| 10/22 | Review | ||
| 10/27 | Midterm | ||
| 10/29 | Midterm Solutions | ||
| 11/3 | Word Sense Disambiguation | bootstrapping | JM 20 |
| 11/5 | Machine Translation | source-channel, model 1 | JM 25 |
| 11/10 | Machine Translation | model 2, 3, hmm | brown 93 |
| 11/12 | Machine Translation | decoding | germann et al. |
| 11/17 | Machine Translation | syntax-based MT | yamada & knight 01 |
| 11/19 | Machine Translation | syntax-based decoding | galley 04, chiang 05 |
| 11/24 | Machine Translation | what's in a translation rule | |
| 12/1 | Machine Translation | tree-to-tree, inversion transduction grammar | |
| 12/3 | Machine Translation | Parameter Optimization, Minimum Bayes Risk | och 03, denero 09 |
| 12/8 | Something fun | Chinese restaurants | goldwater 06 |
| 12/10 | Review | come to class with questions! |