61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video...
Transcript of 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video...
![Page 1: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/1.jpg)
61A Lecture 35
Wednesday, December 4
![Page 2: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/2.jpg)
Announcements
2
![Page 3: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/3.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
2
![Page 4: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/4.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
2
![Page 5: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/5.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
!Come to class and take the final survey.
2
![Page 6: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/6.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
!Come to class and take the final survey.
!There will be a screencast of live lecture (as always).
2
![Page 7: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/7.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
!Come to class and take the final survey.
!There will be a screencast of live lecture (as always).
!Screencasts: http://www.youtube.com/view_play_list?p=-XXv-cvA_iCIEwJhyDVdyLMCiimv6Tup
2
![Page 8: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/8.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
!Come to class and take the final survey.
!There will be a screencast of live lecture (as always).
!Screencasts: http://www.youtube.com/view_play_list?p=-XXv-cvA_iCIEwJhyDVdyLMCiimv6Tup
• Homework 12 due Tuesday 12/10 @ 11:59pm.
2
![Page 9: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/9.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
!Come to class and take the final survey.
!There will be a screencast of live lecture (as always).
!Screencasts: http://www.youtube.com/view_play_list?p=-XXv-cvA_iCIEwJhyDVdyLMCiimv6Tup
• Homework 12 due Tuesday 12/10 @ 11:59pm.
!All you have to do is vote on your favorite recursive art.
2
![Page 10: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/10.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
!Come to class and take the final survey.
!There will be a screencast of live lecture (as always).
!Screencasts: http://www.youtube.com/view_play_list?p=-XXv-cvA_iCIEwJhyDVdyLMCiimv6Tup
• Homework 12 due Tuesday 12/10 @ 11:59pm.
!All you have to do is vote on your favorite recursive art.
• 29 review sessions next week! Come learn about the topics that interest you the most.
2
![Page 11: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/11.jpg)
Announcements
• Homework 11 due Thursday 12/5 @ 11:59pm.
• No video of lecture on Friday 12/6.
!Come to class and take the final survey.
!There will be a screencast of live lecture (as always).
!Screencasts: http://www.youtube.com/view_play_list?p=-XXv-cvA_iCIEwJhyDVdyLMCiimv6Tup
• Homework 12 due Tuesday 12/10 @ 11:59pm.
!All you have to do is vote on your favorite recursive art.
• 29 review sessions next week! Come learn about the topics that interest you the most.
!See http://inst.eecs.berkeley.edu/~cs61a/fa13/exams/final.html for the schedule.
2
![Page 12: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/12.jpg)
Natural Language Processing
![Page 13: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/13.jpg)
Ambiguity in Natural Language
Unlike programming languages, natural languages are ambiguous.
4
![Page 14: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/14.jpg)
Ambiguity in Natural Language
Unlike programming languages, natural languages are ambiguous.
4
Syntactic ambiguity:
![Page 15: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/15.jpg)
Ambiguity in Natural Language
Unlike programming languages, natural languages are ambiguous.
4
Syntactic ambiguity: TEACHER STRIKES IDLE KIDS
![Page 16: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/16.jpg)
Ambiguity in Natural Language
Unlike programming languages, natural languages are ambiguous.
4
Syntactic ambiguity: TEACHER STRIKES IDLE KIDS HOSPITALS ARE SUED BY 7 FOOT DOCTORS
![Page 17: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/17.jpg)
Ambiguity in Natural Language
Unlike programming languages, natural languages are ambiguous.
4
Syntactic ambiguity:
Semantic ambiguity:
TEACHER STRIKES IDLE KIDS HOSPITALS ARE SUED BY 7 FOOT DOCTORS
![Page 18: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/18.jpg)
Ambiguity in Natural Language
Unlike programming languages, natural languages are ambiguous.
4
Syntactic ambiguity:
Semantic ambiguity: IRAQI HEAD SEEKS ARMS
TEACHER STRIKES IDLE KIDS HOSPITALS ARE SUED BY 7 FOOT DOCTORS
![Page 19: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/19.jpg)
Ambiguity in Natural Language
Unlike programming languages, natural languages are ambiguous.
4
Syntactic ambiguity:
Semantic ambiguity: IRAQI HEAD SEEKS ARMS
TEACHER STRIKES IDLE KIDS HOSPITALS ARE SUED BY 7 FOOT DOCTORS
STOLEN PAINTING FOUND BY TREE
![Page 20: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/20.jpg)
Tasks in Natural Language Processing
5
![Page 21: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/21.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
5
![Page 22: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/22.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
5
![Page 23: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/23.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
Machine Translation. "Call a spade a spade!" Google Translate says, "Appeler un chat un chat."
5
![Page 24: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/24.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
Machine Translation. "Call a spade a spade!" Google Translate says, "Appeler un chat un chat."
Semantic Parsing. "When's my birthday?" Siri says, "Your birthday is May 1st."
5
![Page 25: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/25.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
Machine Translation. "Call a spade a spade!" Google Translate says, "Appeler un chat un chat."
Semantic Parsing. "When's my birthday?" Siri says, "Your birthday is May 1st."
Much attention is given to more focused language analysis problems:
5
![Page 26: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/26.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
Machine Translation. "Call a spade a spade!" Google Translate says, "Appeler un chat un chat."
Semantic Parsing. "When's my birthday?" Siri says, "Your birthday is May 1st."
Much attention is given to more focused language analysis problems:
Coreference Resolution: Do the phrases "Barack Obama" and "the president" co-refer?
5
![Page 27: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/27.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
Machine Translation. "Call a spade a spade!" Google Translate says, "Appeler un chat un chat."
Semantic Parsing. "When's my birthday?" Siri says, "Your birthday is May 1st."
Much attention is given to more focused language analysis problems:
Coreference Resolution: Do the phrases "Barack Obama" and "the president" co-refer?
Syntactic Parsing: In "I saw the man with the telescope," who has the telescope?
5
![Page 28: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/28.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
Machine Translation. "Call a spade a spade!" Google Translate says, "Appeler un chat un chat."
Semantic Parsing. "When's my birthday?" Siri says, "Your birthday is May 1st."
Much attention is given to more focused language analysis problems:
Coreference Resolution: Do the phrases "Barack Obama" and "the president" co-refer?
Syntactic Parsing: In "I saw the man with the telescope," who has the telescope?
Word Sense Disambiguation: Does the "bank of the Seine" have an ATM?
5
![Page 29: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/29.jpg)
Tasks in Natural Language Processing
Research in natural language processing (NLP) focuses on tasks that involve language:
Question answering. "Harriet Boyd Hawes was the first woman to discover and excavate a Minoan settlement on this island." Watson says, "What is Crete?"
Machine Translation. "Call a spade a spade!" Google Translate says, "Appeler un chat un chat."
Semantic Parsing. "When's my birthday?" Siri says, "Your birthday is May 1st."
Much attention is given to more focused language analysis problems:
Coreference Resolution: Do the phrases "Barack Obama" and "the president" co-refer?
Syntactic Parsing: In "I saw the man with the telescope," who has the telescope?
Word Sense Disambiguation: Does the "bank of the Seine" have an ATM?
Named-Entity Recognition: What names are in "Did van Gogh paint the Bank of the Seine?"
5
![Page 30: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/30.jpg)
Machine Translation
![Page 31: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/31.jpg)
Machine Translation
7
![Page 32: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/32.jpg)
Machine Translation
Target language corpus gives examples of well-formed sentences
I will get to it later See you later He will do it
7
![Page 33: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/33.jpg)
Machine Translation
Parallel corpus gives translation examples
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
Target language corpus gives examples of well-formed sentences
I will get to it later See you later He will do it
7
![Page 34: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/34.jpg)
Machine Translation
Parallel corpus gives translation examples
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
Machine translation system:
Target language corpus gives examples of well-formed sentences
I will get to it later See you later He will do it
7
![Page 35: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/35.jpg)
Machine Translation
Parallel corpus gives translation examples
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
Machine translation system:
Model of translation
Target language corpus gives examples of well-formed sentences
I will get to it later See you later He will do it
7
![Page 36: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/36.jpg)
Machine Translation
I will do it later
Target language
Parallel corpus gives translation examples
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
Machine translation system:
Model of translation
Target language corpus gives examples of well-formed sentences
I will get to it later See you later He will do it
Yo lo haré despuésNOVEL SENTENCE
Source language
7
![Page 37: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/37.jpg)
Syntactic Agreement in Translation
VB
MD VP
VPNP
S
PRP ADV
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
PRPVB
MD VP
VPNP
S
PRP ADV
I will do it laterModel of translation
Yo lo haré después
Machine translation system:
8
![Page 38: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/38.jpg)
Syntactic Agreement in Translation
VB
MD VP
VPNP
S
PRP ADV
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
PRPVB
MD VP
VPNP
S
PRP ADV
I will do it laterModel of translation
Yo lo haré despuésADV ADV
Machine translation system:
8
![Page 39: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/39.jpg)
Syntactic Agreement in Translation
VB
MD VP
VPNP
S
PRP ADV
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
PRPVB
MD VP
VPNP
S
PRP ADV
I will do it laterModel of translation
Yo lo haré después
S SADV ADV
Machine translation system:
8
![Page 40: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/40.jpg)
Syntactic Agreement in Translation
VB
MD VP
VPNP
S
PRP ADV
Yo lo haré de muy buen grado
I will do it gladly
Después lo veras
You will see later
PRPVB
MD VP
VPNP
S
PRP ADV
I will do it laterModel of translation
Yo lo haré después
S SADV ADV
Machine translation system:
8
![Page 41: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/41.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
![Page 42: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/42.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
![Page 43: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/43.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
pair
NP
S
NN
![Page 44: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/44.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
pair
NP
S
NN
VP
added
VBD
![Page 45: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/45.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
pair
NP
S
NN
VP
added
VBD
![Page 46: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/46.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 47: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/47.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 48: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/48.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 49: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/49.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 50: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/50.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 51: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/51.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
一対 がpair
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 52: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/52.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
一対 がpair
目録list
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 53: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/53.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
一対 がpair
目録list
にto
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 54: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/54.jpg)
Syntactic Reordering in Translation
9
pair added to the lexicon
NP
PP
VP
NP
S
NN VBD TO DT NN
to
PP
TO
一対 がpair
目録list
にto
追加されましたadd was
pair
NP
S
NN
the lexicon
NP
DT NN
VP
added
VBD
![Page 55: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/55.jpg)
Context-Free Grammars
![Page 56: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/56.jpg)
Grammar Rules
A Context-Free Grammar Models Language Generation
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 57: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/57.jpg)
Grammar Rules
A Context-Free Grammar Models Language Generation
S
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 58: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/58.jpg)
S -> NP VP
Grammar Rules
A Context-Free Grammar Models Language Generation
S
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 59: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/59.jpg)
S -> NP VP
Grammar Rules
A Context-Free Grammar Models Language Generation
S
NP VP
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 60: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/60.jpg)
S -> NP VP
NP -> PRP
Grammar Rules
A Context-Free Grammar Models Language Generation
S
NP VP
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 61: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/61.jpg)
S -> NP VP
NP -> PRP
Grammar Rules
A Context-Free Grammar Models Language Generation
S
NP VP
PRP
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 62: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/62.jpg)
S -> NP VP
NP -> PRP
Grammar Rules
Lexicon
A Context-Free Grammar Models Language Generation
S
NP VP
PRP
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 63: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/63.jpg)
S -> NP VP
NP -> PRP
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 64: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/64.jpg)
S -> NP VP
NP -> PRP
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP
I
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 65: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/65.jpg)
S -> NP VP
NP -> PRP
VP -> VB
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP
I
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 66: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/66.jpg)
S -> NP VP
NP -> PRP
VP -> VB
VP -> VB NP
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP
I
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 67: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/67.jpg)
S -> NP VP
NP -> PRP
VP -> VB
VP -> VB NP
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP VB NP
I
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 68: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/68.jpg)
S -> NP VP
NP -> PRP
VP -> VB
VP -> VB NP
VB -> know
VB -> help
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP VB NP
I
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 69: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/69.jpg)
S -> NP VP
NP -> PRP
VP -> VB
VP -> VB NP
VB -> know
VB -> help
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP VB NP
I know
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 70: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/70.jpg)
S -> NP VP
NP -> PRP
VP -> VB
VP -> VB NP
VB -> know
VB -> help
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP VB NP
I know PRP
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 71: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/71.jpg)
S -> NP VP
NP -> PRP
VP -> VB
VP -> VB NP
PRP -> you
VB -> know
VB -> help
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP VB NP
I know PRP
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 72: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/72.jpg)
S -> NP VP
NP -> PRP
VP -> VB
VP -> VB NP
PRP -> you
VB -> know
VB -> help
Grammar Rules
Lexicon
PRP -> I
A Context-Free Grammar Models Language Generation
S
NP VP
PRP VB NP
I know PRP
you
A grammar contains rules that hierarchically generate word sequences using syntactic tags.
11
![Page 73: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/73.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VB -> know
VB -> help
Grammar Rules
Lexicon
S
NP VP
PRP
I
12
![Page 74: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/74.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VB -> know
VB -> help
Grammar Rules
Lexicon
S
NP VP
PRP
I
12
![Page 75: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/75.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
Grammar Rules
Lexicon
S
NP VP
PRP
I
12
![Page 76: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/76.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
Grammar Rules
Lexicon
S
NP VP
PRP
I
12
![Page 77: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/77.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
Grammar Rules
Lexicon
S
NP VP
PRP
I
0.2
0.7
0.1
12
![Page 78: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/78.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
Grammar Rules
Lexicon
S
NP VP
PRP
I
MD VP
0.2
0.7
0.1
12
![Page 79: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/79.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
MD -> can
Grammar Rules
Lexicon
S
NP VP
PRP
I
MD VP
0.2
0.7
0.1
12
![Page 80: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/80.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
MD -> can
Grammar Rules
Lexicon
S
NP VP
PRP
I can
MD VP
0.2
0.7
0.1
12
![Page 81: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/81.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
MD -> can
Grammar Rules
Lexicon
S
NP VP
PRP
I can
MD VP
0.2
0.7
0.1
12
![Page 82: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/82.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
MD -> can
Grammar Rules
Lexicon
S
NP VP
PRP
I can
MD VP
VB NP
help
0.2
0.7
0.1
12
![Page 83: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/83.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
MD -> can
Grammar Rules
Lexicon
S
NP VP
PRP
I can
MD VP
VB NP
help PRP
0.2
0.7
0.1
12
![Page 84: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/84.jpg)
Probabilistic Context-Free Grammars
S -> NP VP
NP -> PRP
PRP -> I
VP -> VB
VP -> VB NP
PRP -> you
VP -> MD VP
VB -> know
VB -> help
MD -> can
Grammar Rules
Lexicon
S
NP VP
PRP
I can
MD VP
VB NP
help PRP
you
0.2
0.7
0.1
12
![Page 85: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/85.jpg)
Learning Probabilistic Context-Free Grammars
(Demo)
![Page 86: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/86.jpg)
Parsing with Probabilistic Context-Free Grammars
![Page 87: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/87.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
15
![Page 88: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/88.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
time flies like an arrow
15
![Page 89: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/89.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
time flies like an arrow
15
![Page 90: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/90.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
time flies like an arrow
15
![Page 91: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/91.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananastime flies like an arrow
15
![Page 92: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/92.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananastime flies like an arrow
15
![Page 93: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/93.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananastime flies like an arrow
15
![Page 94: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/94.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow
15
![Page 95: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/95.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
15
![Page 96: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/96.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
15
![Page 97: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/97.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
S -> NP VP
15
![Page 98: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/98.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
S -> NP VP
NP -> NN
15
![Page 99: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/99.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time
S -> NP VP
NP -> NN
15
![Page 100: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/100.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time
S -> NP VP
NP -> NN VP -> VBZ PP
15
![Page 101: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/101.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time VBZ -> flies
S -> NP VP
NP -> NN VP -> VBZ PP
15
![Page 102: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/102.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time VBZ -> flies
S -> NP VP
NP -> NN VP -> VBZ PP
PP -> IN NP
15
![Page 103: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/103.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time VBZ -> flies IN -> like
S -> NP VP
NP -> NN VP -> VBZ PP
PP -> IN NP
15
![Page 104: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/104.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time VBZ -> flies IN -> like
S -> NP VP
NP -> NN VP -> VBZ PP
PP -> IN NP
NP -> DT NN
15
![Page 105: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/105.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time VBZ -> flies IN -> like DT -> an
S -> NP VP
NP -> NN VP -> VBZ PP
PP -> IN NP
NP -> DT NN
15
![Page 106: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/106.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time VBZ -> flies IN -> like DT -> an NN -> arrow
S -> NP VP
NP -> NN VP -> VBZ PP
PP -> IN NP
NP -> DT NN
15
![Page 107: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/107.jpg)
Parsing is Maximizing Likelihood
A probabilistic context-free grammar can be used to select a parse for a sentence.
fruit flies like bananas
Parse by finding the tree with the highest total probability that yields the sentence.
time flies like an arrow0 1 2 3 4 5
time flies like an arrow
Algorithm: Try every rule over every span. Match the lexicon to each word.
NN -> time VBZ -> flies IN -> like DT -> an NN -> arrow
S -> NP VP
NP -> NN VP -> VBZ PP
PP -> IN NP
NP -> DT NN
(Demo)
15
![Page 108: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/108.jpg)
Tree Transformations
![Page 109: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/109.jpg)
Reordering Modal Arguments
17
![Page 110: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/110.jpg)
Reordering Modal Arguments
English
17
![Page 111: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/111.jpg)
Reordering Modal Arguments
English Yoda-English
17
![Page 112: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/112.jpg)
Reordering Modal Arguments
English Yoda-English Help you, I can! Yes! Mm!
17
![Page 113: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/113.jpg)
Reordering Modal Arguments
English Yoda-English Help you, I can! Yes! Mm!
When 900 years old you reach, look as good, you will not. Hm.
17
![Page 114: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/114.jpg)
Reordering Modal Arguments
English Yoda-English Help you, I can! Yes! Mm!
When 900 years old you reach, look as good, you will not. Hm.
S
NP VP
PRP
I can
MD VP
VB PRP
help you
17
![Page 115: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/115.jpg)
Reordering Modal Arguments
English Yoda-English Help you, I can! Yes! Mm!
When 900 years old you reach, look as good, you will not. Hm.
S
NP VP
PRP
I can
MD VP
VB PRP
help you
17
![Page 116: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/116.jpg)
Reordering Modal Arguments
English Yoda-English Help you, I can! Yes! Mm!
When 900 years old you reach, look as good, you will not. Hm.
S
NP VP
PRP
I can
MDVB PRP
help you
VP
17
![Page 117: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/117.jpg)
Reordering Modal Arguments
English Yoda-English Help you, I can! Yes! Mm!
When 900 years old you reach, look as good, you will not. Hm.
S
NP VP
PRP
I can
MDVB PRP
help you
VP .
,
17
![Page 118: 61A Lecture 35 - University of California, Berkeleycs61a/fa13/slides/35-NLP_1pp.pdf · •No video of lecture on Friday 12/6.!Come to class and take the final survey. 2. Announcements](https://reader034.fdocuments.net/reader034/viewer/2022043018/5f3a4399b37012096e06dfd8/html5/thumbnails/118.jpg)
Reordering Modal Arguments
English Yoda-English Help you, I can! Yes! Mm!
When 900 years old you reach, look as good, you will not. Hm.
S
NP VP
PRP
I can
MDVB PRP
help you
VP .
,
(Demo)
17