parallel tools for natural language processing
DESCRIPTION
Parallel Tools for Natural Language Processing. Mark Brigham Melanie Goetz Andrew Hogue. 6.338 / 18.337 - March 16, 2004. Sentence Parsing. Consider the sentence: “John ate the cookie on the table” We want to: Tag the sentence with parts of speech Group the words by phrase. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/1.jpg)
Parallel Tools for Natural Language Processing
Mark Brigham
Melanie Goetz
Andrew Hogue
6.338 / 18.337 - March 16, 2004
![Page 2: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/2.jpg)
• Consider the sentence:
“John ate the cookie on the table”
• We want to:– Tag the sentence with parts of speech– Group the words by phrase
Sentence Parsing
![Page 3: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/3.jpg)
…
Context Free Grammars
• Recursive set of rules
• Defines what syntactic structure can be applied to a phrase or word
• Top-level rule S defines the sentence
S → NP VP
NP → Det N
NP → NP PP
VP → VP PP
VP → V NP
N → ‘cookie’
N → ‘table’
Det → ‘the’
V → ‘ate’
![Page 4: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/4.jpg)
Context Free Grammars
• Applying a CFG to a sentence creates a parse-tree for that sentence
![Page 5: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/5.jpg)
Context Free Grammars
Top-downparse
![Page 6: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/6.jpg)
Context Free Grammars
Bottom-upparse
Parallelizable!
![Page 7: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/7.jpg)
Ambiguity
More than one parse for a single sentence!
![Page 8: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/8.jpg)
Parallelization
• Bottom-up rule application appropriate for parallel processing
• Ambiguous parses also parallelizable
• Long, complex sentences may be most interesting
• Proust?
![Page 9: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/9.jpg)
Chart Parsing
• Create a matrix where entries correspond to words/phrases
• If there is a valid CFG parse of a phrase [i,j], add it to that matrix cell
• A cell [i,j] may only depend on other cells [m,n] where m < i and n < j.
![Page 10: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/10.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 11: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/11.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 12: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/12.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 13: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/13.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 14: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/14.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 15: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/15.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 16: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/16.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 17: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/17.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 18: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/18.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 19: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/19.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 20: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/20.jpg)
John ate the cookie on the table
John
ate
the
cookie
on
the
table
![Page 21: Parallel Tools for Natural Language Processing](https://reader035.vdocuments.net/reader035/viewer/2022081519/5681351a550346895d9c6f4d/html5/thumbnails/21.jpg)
Other Tools
• Considering parallelizing other NLP tools
• Word-stemming: Multiple finite state automata applied to a single word in parallel
• Automated part-of-speech recognition on large corpora