schliessen

Filtern

 

Bibliotheken

Cover Image

Parallel Parsing of Tree Adjoining Grammars on the Connection Machine

A parsing algorithm for a tree adjoining grammar (TAG) and its parallel implementation on the Connection Machine (CM) are described. A TAG is a formalism for natural language that employs trees as the basic grammar structures. Parsing involves the application of 2 operations, called adjunction and s... Full description

Journal Title: International journal of parallel programming 1992, Vol.21 (1), p.1-38
Main Author: Palis, Michael A
Other Authors: Wei, David S. L
Format: Electronic Article Electronic Article
Language: English
Subjects:
990200 - Mathematics & Computers
ALGORITHMS
Applied sciences
computational linguistics
COMPUTER ARCHITECTURE
Computer programming
Computer science
control theory
systems
Exact sciences and technology
GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
Grammar
Implementations
language
Language theory and syntactical analysis
mathematical linguistics
MATHEMATICAL LOGIC
Natural language
Natural language interfaces
OPTIMIZATION
parallel parsing
PARALLEL PROCESSING
Parsing
PROGRAMMING
Theoretical computing
Theory
tree adjoining grammar
Usage
Publisher: New York, NY: Plenum Press
ID: ISSN: 0885-7458
Link: http://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=4753645
Zum Text:
SendSend as email Add to Book BagAdd to Book Bag
Staff View
recordid: cdi_osti_scitechconnect_7249394
title: Parallel Parsing of Tree Adjoining Grammars on the Connection Machine
format: Article
creator:
  • Palis, Michael A
  • Wei, David S. L
subjects:
  • 990200 - Mathematics & Computers
  • ALGORITHMS
  • Applied sciences
  • computational linguistics
  • COMPUTER ARCHITECTURE
  • Computer programming
  • Computer science
  • control theory
  • systems
  • Exact sciences and technology
  • GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
  • Grammar
  • Implementations
  • language
  • Language theory and syntactical analysis
  • mathematical linguistics
  • MATHEMATICAL LOGIC
  • Natural language
  • Natural language interfaces
  • OPTIMIZATION
  • parallel parsing
  • PARALLEL PROCESSING
  • Parsing
  • PROGRAMMING
  • Theoretical computing
  • Theory
  • tree adjoining grammar
  • Usage
ispartof: International journal of parallel programming, 1992, Vol.21 (1), p.1-38
description: A parsing algorithm for a tree adjoining grammar (TAG) and its parallel implementation on the Connection Machine (CM) are described. A TAG is a formalism for natural language that employs trees as the basic grammar structures. Parsing involves the application of 2 operations, called adjunction and substitution, to produce derived tree structures. Sequential parsing algorithms for TAGs run in time quadratic in the grammar size, which is impractical for the very large grammars currently being developed for natural language. Two parallel algorithms, one running in time nearly linear in the grammar size and the other running in time logarithmic in the grammar size, are presented. Both parallel algorithms have been implemented on a CM, and performance measurements were obtained for varying grammar sizes. For the range of grammar sizes encountered in real applications, the first parallel algorithm appears to be more suitable for current versions of the CM. The 2nd algorithm is perhaps more suitable for future versions of the CM.
language: eng
source:
identifier: ISSN: 0885-7458
fulltext: no_fulltext
issn:
  • 0885-7458
  • 1573-7640
url: Link


@attributes
NO1
SEARCH_ENGINEprimo_central_multiple_fe
SEARCH_ENGINE_TYPEPrimo Central Search Engine
RANK1.4689049
LOCALfalse
PrimoNMBib
record
control
sourceidgale_osti_
recordidTN_cdi_osti_scitechconnect_7249394
sourceformatXML
sourcesystemPC
galeidA13830506
ch_idR02866904
sourcerecordidA13830506
originalsourceidFETCH-LOGICAL-c391t-ed2c55d2f068528348034467ba759680b868586b311f639de4b3278d448fba4b0
addsrcrecordideNptkV9rFDEUxYMouFZf_ASDKIIw9Wby_3Fd2ipUFKnPIZNJOllnk5rMUvbbN-suXR9KHpKc_O7h3FyE3mI4xwDi85dLwEQogskztMBMkFZwCs_RAqRkraBMvkSvSlkDgBJSLtDFT5PNNLmpqYcS4m2TfHOTnWuWwzqFuFeustls6muTYjOPrlmlGJ2dQ71-N3YM0b1GL7yZintz3M_Q78uLm9XX9vrH1bfV8rq1ROG5dUNnGRs6D1yyThIqgVDKRW8EU1xCL6sueU8w9pyowdGedEIOlErfG9rDGXp38E1lDrrYMDs72kMcLTqqiKIV-niA7nL6u3Vl1ptQrJsmE13aFi0oIVJSIk52j-Q6bXOsHegOaKdAAT9Bt2ZyOkSf5mzs3lIvMZEE2D-ofQKqa3CbUAM6H6r-P__pwNucSsnO67sc6h_vNAa9n6Q-TbLC748xTbFm8tlEG8pjBRWMcMoq9uHoOZrhfmf_jK6G2WnT1971L-gk5wooPADCFKQI
sourcetypeOpen Access Repository
isCDItrue
recordtypearticle
pqid204290906
display
typearticle
titleParallel Parsing of Tree Adjoining Grammars on the Connection Machine
creatorPalis, Michael A ; Wei, David S. L
creatorcontribPalis, Michael A ; Wei, David S. L
descriptionA parsing algorithm for a tree adjoining grammar (TAG) and its parallel implementation on the Connection Machine (CM) are described. A TAG is a formalism for natural language that employs trees as the basic grammar structures. Parsing involves the application of 2 operations, called adjunction and substitution, to produce derived tree structures. Sequential parsing algorithms for TAGs run in time quadratic in the grammar size, which is impractical for the very large grammars currently being developed for natural language. Two parallel algorithms, one running in time nearly linear in the grammar size and the other running in time logarithmic in the grammar size, are presented. Both parallel algorithms have been implemented on a CM, and performance measurements were obtained for varying grammar sizes. For the range of grammar sizes encountered in real applications, the first parallel algorithm appears to be more suitable for current versions of the CM. The 2nd algorithm is perhaps more suitable for future versions of the CM.
identifier
0ISSN: 0885-7458
1EISSN: 1573-7640
2DOI: 10.1007/BF01379313
3CODEN: IJPPE5
languageeng
publisherNew York, NY: Plenum Press
subject990200 - Mathematics & Computers ; ALGORITHMS ; Applied sciences ; computational linguistics ; COMPUTER ARCHITECTURE ; Computer programming ; Computer science; control theory; systems ; Exact sciences and technology ; GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE ; Grammar ; Implementations ; language ; Language theory and syntactical analysis ; mathematical linguistics ; MATHEMATICAL LOGIC ; Natural language ; Natural language interfaces ; OPTIMIZATION ; parallel parsing ; PARALLEL PROCESSING ; Parsing ; PROGRAMMING ; Theoretical computing ; Theory ; tree adjoining grammar ; Usage
ispartofInternational journal of parallel programming, 1992, Vol.21 (1), p.1-38
rights
01993 INIST-CNRS
1Copyright Plenum Publishing Corporation Feb 1992
lds50peer_reviewed
citesFETCH-LOGICAL-c391t-ed2c55d2f068528348034467ba759680b868586b311f639de4b3278d448fba4b0
links
openurl$$Topenurl_article
thumbnail$$Usyndetics_thumb_exl
backlink$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=4753645$$DView record in Pascal Francis
search
creatorcontrib
0Palis, Michael A
1Wei, David S. L
title
0Parallel Parsing of Tree Adjoining Grammars on the Connection Machine
1International journal of parallel programming
descriptionA parsing algorithm for a tree adjoining grammar (TAG) and its parallel implementation on the Connection Machine (CM) are described. A TAG is a formalism for natural language that employs trees as the basic grammar structures. Parsing involves the application of 2 operations, called adjunction and substitution, to produce derived tree structures. Sequential parsing algorithms for TAGs run in time quadratic in the grammar size, which is impractical for the very large grammars currently being developed for natural language. Two parallel algorithms, one running in time nearly linear in the grammar size and the other running in time logarithmic in the grammar size, are presented. Both parallel algorithms have been implemented on a CM, and performance measurements were obtained for varying grammar sizes. For the range of grammar sizes encountered in real applications, the first parallel algorithm appears to be more suitable for current versions of the CM. The 2nd algorithm is perhaps more suitable for future versions of the CM.
subject
0990200 - Mathematics & Computers
1ALGORITHMS
2Applied sciences
3computational linguistics
4COMPUTER ARCHITECTURE
5Computer programming
6Computer science; control theory; systems
7Exact sciences and technology
8GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
9Grammar
10Implementations
11language
12Language theory and syntactical analysis
13mathematical linguistics
14MATHEMATICAL LOGIC
15Natural language
16Natural language interfaces
17OPTIMIZATION
18parallel parsing
19PARALLEL PROCESSING
20Parsing
21PROGRAMMING
22Theoretical computing
23Theory
24tree adjoining grammar
25Usage
issn
00885-7458
11573-7640
fulltextfalse
rsrctypearticle
creationdate1992
recordtypearticle
recordideNptkV9rFDEUxYMouFZf_ASDKIIw9Wby_3Fd2ipUFKnPIZNJOllnk5rMUvbbN-suXR9KHpKc_O7h3FyE3mI4xwDi85dLwEQogskztMBMkFZwCs_RAqRkraBMvkSvSlkDgBJSLtDFT5PNNLmpqYcS4m2TfHOTnWuWwzqFuFeustls6muTYjOPrlmlGJ2dQ71-N3YM0b1GL7yZintz3M_Q78uLm9XX9vrH1bfV8rq1ROG5dUNnGRs6D1yyThIqgVDKRW8EU1xCL6sueU8w9pyowdGedEIOlErfG9rDGXp38E1lDrrYMDs72kMcLTqqiKIV-niA7nL6u3Vl1ptQrJsmE13aFi0oIVJSIk52j-Q6bXOsHegOaKdAAT9Bt2ZyOkSf5mzs3lIvMZEE2D-ofQKqa3CbUAM6H6r-P__pwNucSsnO67sc6h_vNAa9n6Q-TbLC748xTbFm8tlEG8pjBRWMcMoq9uHoOZrhfmf_jK6G2WnT1971L-gk5wooPADCFKQI
startdate1992
enddate1992
creator
0Palis, Michael A
1Wei, David S. L
general
0Plenum Press
1Springer
2Springer Nature B.V
scope
0CBS
1IQODW
2AAYXX
3CITATION
4BSHEE
57SC
68FD
7JQ2
8L7M
9L~C
10L~D
11OTOTI
sort
creationdate1992
titleParallel Parsing of Tree Adjoining Grammars on the Connection Machine
authorPalis, Michael A ; Wei, David S. L
facets
frbrtype5
frbrgroupidcdi_FETCH-LOGICAL-c391t-ed2c55d2f068528348034467ba759680b868586b311f639de4b3278d448fba4b0
rsrctypearticles
prefilterarticles
languageeng
creationdate1992
topic
0990200 - Mathematics & Computers
1ALGORITHMS
2Applied sciences
3computational linguistics
4COMPUTER ARCHITECTURE
5Computer programming
6Computer science; control theory; systems
7Exact sciences and technology
8GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
9Grammar
10Implementations
11language
12Language theory and syntactical analysis
13mathematical linguistics
14MATHEMATICAL LOGIC
15Natural language
16Natural language interfaces
17OPTIMIZATION
18parallel parsing
19PARALLEL PROCESSING
20Parsing
21PROGRAMMING
22Theoretical computing
23Theory
24tree adjoining grammar
25Usage
toplevelpeer_reviewed
creatorcontrib
0Palis, Michael A
1Wei, David S. L
collection
0Annual Bibliography of English Language and Literature
1Pascal-Francis
2CrossRef
3Academic OneFile (A&I only)
4Computer and Information Systems Abstracts
5Technology Research Database
6ProQuest Computer Science Collection
7Advanced Technologies Database with Aerospace
8Computer and Information Systems Abstracts – Academic
9Computer and Information Systems Abstracts Professional
10OSTI.GOV
jtitleInternational journal of parallel programming
delivery
delcategoryRemote Search Resource
fulltextno_fulltext
addata
au
0Palis, Michael A
1Wei, David S. L
formatjournal
genrearticle
ristypeJOUR
atitleParallel Parsing of Tree Adjoining Grammars on the Connection Machine
jtitleInternational journal of parallel programming
date1992
risdate1992
volume21
issue1
spage1
epage38
pages1-38
issn0885-7458
eissn1573-7640
codenIJPPE5
abstractA parsing algorithm for a tree adjoining grammar (TAG) and its parallel implementation on the Connection Machine (CM) are described. A TAG is a formalism for natural language that employs trees as the basic grammar structures. Parsing involves the application of 2 operations, called adjunction and substitution, to produce derived tree structures. Sequential parsing algorithms for TAGs run in time quadratic in the grammar size, which is impractical for the very large grammars currently being developed for natural language. Two parallel algorithms, one running in time nearly linear in the grammar size and the other running in time logarithmic in the grammar size, are presented. Both parallel algorithms have been implemented on a CM, and performance measurements were obtained for varying grammar sizes. For the range of grammar sizes encountered in real applications, the first parallel algorithm appears to be more suitable for current versions of the CM. The 2nd algorithm is perhaps more suitable for future versions of the CM.
copNew York, NY
pubPlenum Press
doi10.1007/BF01379313