chapter 13 natural language processing. machine translation

23
CHAPTER 13 CHAPTER 13 NATURAL LANGUAGE PROCESSING

Upload: phyllis-hodges

Post on 02-Jan-2016

236 views

Category:

Documents


4 download

TRANSCRIPT

Page 1: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

CHAPTER 13CHAPTER 13

NATURAL LANGUAGE PROCESSING

Page 2: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Machine Translation

Page 3: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Information Extraction

Page 4: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Question Answering

Page 5: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Some Early NLP History

Page 6: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Why is Language Hard?• Ambiguity

Eye Drops Off Shelf Miners Refuse to Work After Death Killer Sentenced to Die for Second Time in 1

Years Iraqi Head Seeks Arms Ban on Nude Dancing on Governor’s Desk Juvenile Court to Try Shooting Defendant Stolen Painting Found by Tree Local HS Dropouts Cut in Half Hospitals Are Sued by 7 Foot Doctors

Page 7: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Models of Language

Page 8: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Language Modeling

Page 9: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Unigram ModelsUnigram Models

Page 10: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Bigram Models

Page 11: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Human Processing

Page 12: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Why is Language Hard?

Page 13: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Parsing as Search: Top-Down

Page 14: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Corpus-Based Methods

Page 15: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Semantic Interpretation• On to meaning!

A very basic approach to computational semantics Truth-theoretic notion of semantics (Tarskian) Assign a “meaning” to each word Word meanings combine according to the parse

structure People can and do spend entire courses on this

topic We’ll spend under an hour!

• What’s NLP and what’s general AI? Designing meaning representations? Computing those representations? Reasoning with them?

Page 16: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Problem: Ambiguities

• Headlines: Iraqi Head Seeks Arms Ban on Nude Dancing on Governor’s Desk Juvenile Court to Try Shooting Defendant Teacher Strikes Idle Kids Stolen Painting Found by Tree Kids Make Nutritious Snacks Local HS Dropouts Cut in Half Hospitals Are Sued by 7 Foot Doctors

• Why are these funny?

Page 17: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Machine Translation

Page 18: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Just a Code?

• “Also knowing nothing official about, but having guessed andinferred considerable about, the powerful new mechanizedmethods in cryptography—methods which I believe succeedeven when one does not know what language has beencoded—one naturally wonders if the problem of translationcould conceivably be treated as a problem in cryptography.When I look at an article in Russian, I say: ‘This is reallywritten in English, but it has been coded in some strangesymbols. I will now proceed to decode.’ ”

• Warren Weaver (1955:18, quoting a letter he wrote in 1947)

Page 19: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Levels of Transfer

Page 20: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Memory: Theory

Page 21: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Time: Theory

Page 22: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Problem: ScaleProblem: Scale

Page 23: CHAPTER 13 NATURAL LANGUAGE PROCESSING. Machine Translation

Problem: Sparsity