Unsupervised!eneratio" of #ara$el %reebanks through &ub'%ree (lignmen)
|
|
|
- Walter Sullivan
- 5 years ago
- Views:
Transcription
1 Unsupervised eneratio" of #ara$el %reebanks through &ub'%ree (lignmen) Ventsislav Zhechev
2 %alk *utlin+ What is a Parallel Treebank and Why do we Need One? Syem Design and Features Software Package Details Avenues for Improvement Conclusions 2/ 30
3 ,hat - a #ara$el %reebank? 3/ 30
4 ,hat - a #ara$el %reebank? English I do not think it is necessary for classic cars to be part of the dire#ive. I am not looking for such rigidly high recycling quotas when it comes to special"purpose vehicles either. I want special"purpose vehicles such as ambulances to have high recovery quotas. This is my main concern in this matter. German Ich halte es nicht für notwendig, daß Oldtimer Beandteil dieser Richtlinie sind. Auch bei Sonderfahrzeugen rebe ich nicht so unbedingt hohe Recyclingquoten an. Ich habe den Wunsch, daß Sonderfahrzeuge wie Krankenwagen hohe Rettungsquoten haben. Das ist meine Hauptsorge in diesem Bereich. 3/ 30
5 / 30,hat - a #ara$el %reebank? 3 English German S NP DT This VP VBZ is NP PRP$ my ADJP JJ main NN concern PP IN in NP DT this NN matter.. TOP S PDS Das VAFIN ist NP PPOSAT meine NN Hauptsorge PP APPR in PDAT diesem NN Bereich $..
6 / 30,hat - a #ara$el %reebank? 3 English German S NP DT This VP VBZ is NP PRP$ my ADJP JJ main NN concern PP IN in NP DT this NN matter.. TOP S PDS Das VAFIN ist NP PPOSAT meine NN Hauptsorge PP APPR in PDAT diesem NN Bereich $..
7 / 30,hat - a #ara$el %reebank? 3 English German S NP DT This VP VBZ is NP PRP$ my ADJP JJ main NN concern PP IN in NP DT this NN matter.. TOP S PDS Das VAFIN ist NP PPOSAT meine NN Hauptsorge PP APPR in PDAT diesem NN Bereich $..
8 / 30,hat - a #ara$el %reebank? 3 English German S NP DT This VP VBZ is NP PRP$ my ADJP JJ main NN concern PP IN in NP DT this NN matter.. TOP S PDS Das VAFIN ist NP PPOSAT meine NN Hauptsorge PP APPR in PDAT diesem NN Bereich $..
9 Us. of #ara$el %reebanks 4/ 30
10 Us. of #ara$el %reebanks Hiero$ $ $ $ $ $ %Chiang, 2007& Probabiliic Synchronous Tree"Insertion Grammars$ $ $ $ %Nesson et al., 2006& Data"Oriented Translation $ $ $ $ $ $ $ $ %Hearne and Way, 2006& Stat"XFER$$ $ $ %Lavie, 2008& 4/ 30
11 Us. of #ara$el %reebanks English NP DT This VBZ is Hiero$ $ $ $ $ $ %Chiang, 2007& S German Probabiliic Synchronous Tree"Insertion VP Grammars$ $ $ $ %Nesson. et al., 2006& Data"Oriented NP Translation PP PRP$ $ $ $ $ ADJP IN NP $ $ $ $ %Hearne and Way, 2006& my in JJ main NN concern DT this NN matter Stat"XFER$$ $ $ %Lavie, 2008& S TOP $.. PDS Das VAFIN ist PPOSAT meine NP NN Hauptsorge APPR in PP PDAT diesem NN Bereich 4/ 30
12 Us. of #ara$el %reebanks PRP$ my Hiero$ $ $ $ $ $ %Chiang, PP 2007& Probabiliic Synchronous DT Tree"Insertion NN this matter Grammars$ NP $ $ $ %Nesson probability et al., 2006& 1 ADJP Data"Oriented JJ NN Translation main probability concern 2 $ $ $ $ $ $ $ $ NP %Hearne and Way, 2006& PPOSAT Stat"XFER$$ meine $ $ %Lavie, Hauptsorge 2008& ADJP IN in NN NP JJ main NN concern APPR in probability 3 PP PDAT diesem NN Bereich NN Hauptsorge 4/ 30
13 ,hat do we need to generat+ #ara$el %reebanks? 6/ 30
14 ,hat do we need to generat+ #ara$el %reebanks? We already have: Parallel Corpora Parsers What s missing? A Sub"Tree Aligner for exiing research see $ %Tinsley et al., 2007&, MT Summit XI $ %Zhechev, to appear&, PhD Thesis 6/ 30
15 ,hat do we wa/ in a &ub'%ree (ligner? 7/ 30
16 ,hat do we wa/ in a &ub'%ree (ligner? Independence Preservation Minimal External Resources Guided Lexical Alignments 7/ 30
17 0ow do. th+ &ub'%ree (ligner,ork? 8/ 30
18 0ow do. th+ &ub'%ree (ligner,ork? Prerequisites: sentence"aligned parallel text monolingual parsers for both languages source"to"target and target"to"source word"alignment probability tables Both sides of the parallel text need to be parsed in advance The sub"tree aligner operates on one parsed sentence pair at a time 8/ 30
19 0ow do. the &ub'%ree (ligner,ork? Initialisation extra# the relevant word"alignment data calculate scores for all possible links between nodes in the source and target tree only nonzero scores are ored Sele#ion sele# the be combination of links for the sentence pair two sele#ion algorithms 9/ 30
20 %ra1lational 2quivalenc+ 10/ 30
21 %ra1lational 2quivalenc+ a b c w z x y inside s l = b c t l = x y outside s l = a t l = w z s, t = " s l t l #" t l s l #" s l t l #" t l s l score1 x # i " j y x = P y j x i y 10/ 30
22 %ra1lational 2quivalenc+ a b c w z x y inside s l = b c t l = x y outside s l = a t l = w z s, t = " s l t l #" t l s l #" s l t l #" t l s l score2 " i y x = P y j x i y # j x P y j x i x 10/ 30
23 reedy'&earch &ele3io" 11/ 30
24 reedy'&earch &ele3io" while unprocessed hypotheses remain do link the highest-scoring hypothesis discard all incompatible hypotheses end while 11/ 30
25 reedy'&earch &ele3io" while unprocessed hypotheses remain do link the highest-scoring hypothesis discard all incompatible hypotheses end while Two hypotheses are incompatible if they share a node 11/ 30
26 reedy'&earch &ele3io" while unprocessed hypotheses remain do link the highest-scoring hypothesis discard all incompatible hypotheses end while Two hypotheses are incompatible if they share a node a w z b c x y 11/ 30
27 reedy'&earch &ele3io" while unprocessed hypotheses remain do link the highest-scoring hypothesis discard all incompatible hypotheses end while Two hypotheses are incompatible if they share a node descendants / anceors of the source node are linked to non"descendants / non"anceors of the target node 11/ 30
28 reedy'&earch &ele3io" while unprocessed hypotheses remain do link the highest-scoring hypothesis discard all incompatible hypotheses end while Two hypotheses are incompatible if they share a node descendants a / anceors w of the source z node are linked b c x y to non"descendants / non"anceors of the target node 11/ 30
29 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" COLON11 : / 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes descendants / anceors of the source node are linked to non"descendants / non"anceors of the target node COLON9 :
30 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" COLON11 : 12/ 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes 10 8 score_1 4 3 score_2 1 1 score_3 8 8 score_ score_5 2 2 score_6 9 5 score_7 1 2 score_8 2 1 score_9 9 9 score_10 descendants / anceors of the source node are linked to non"descendants / non"anceors of the target node '.e"() *.e"'+ COLON9 :
31 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" 1 1 COLON11 : 12/ 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes 10 8 score_1 4 3 score_2 1 1 score_3 8 8 score_ score_5 2 2 score_6 9 5 score_7 1 2 score_8 2 1 score_9 9 9 score_10 descendants / anceors of the source node are linked to non"descendants / non"anceors of the target node COLON9 :
32 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" 1 1 COLON11 : 12/ 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes 4 3 score_2 1 1 score_ score_5 2 2 score_6 9 5 score_7 1 2 score_8 2 1 score_9 9 9 score_10 descendants / anceors of the source node are 2 linked to non"descendants / non"anceors of the target node COLON9 :
33 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" 1 1 COLON11 : 12/ 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes 1 1 score_ score_5 2 2 score_6 9 5 score_7 1 2 score_8 2 1 score_9 9 9 score_10 descendants / anceors of the source node are 2 linked to non"descendants / non"anceors of the target node COLON9 :
34 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" COLON11 : 12/ 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes descendants / anceors of the source node are 2 linked to non"descendants / non"anceors of the target node 11 9 score_5 2 2 score_6 9 5 score_7 9 9 score_10 COLON9 :
35 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" COLON11 : 12/ 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes descendants / anceors of the source node are 2 linked to non"descendants / non"anceors of the target node 2 2 score_6 9 5 score_7 COLON9 :
36 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" COLON11 : 12/ 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes descendants / anceors of the source node are 2 linked to non"descendants / non"anceors of the target node 9 5 score_7 COLON9 :
37 ROOT1 ROOT1 V4 check V3 CONJ6 or VPv2 while unprocessed hypotheses remain do link the NP8 highest-scoring V3 hypothesis NPdet4 discard D9 all N10 incompatible effectuez hypotheses V7 end while the following D5 NPap6 do reedy'&earch &ele3io" COLON11 : are linked to 4 non"descendants 2 / non"anceors of the target node / 30 VPverb main 2 les N7 Two hypotheses are incompatible if they share a node vérifications A8 suivantes descendants / anceors of the source node COLON9 :
38 reedy'&earch &ele3io" while unprocessed hypotheses remain do link the highest-scoring hypothesis discard all incompatible hypotheses end while Two hypotheses are incompatible if they share a node descendants / anceors of the source node are linked to non"descendants / non"anceors of the target node Several hypotheses may share the highe translational equivalence score 12/ 30
39 skip2 skip1 while unprocessed hypotheses with no tied competitors remain do while the highest-scoring hypothesis has tied competitors do skip all tied competitors end while link the highest-scoring hypothesis discard all incompatible hypotheses end while a b a c d c e f g b h i score_1 score_1 score_2 score_3 score_4 score_4 13/ 30
40 skip2 skip1 while unprocessed hypotheses with no tied competitors remain do while the highest-scoring hypothesis has tied competitors do skip all tied competitors end while link the highest-scoring hypothesis discard all incompatible hypotheses end while a b a c d c e f g b h i score_1 score_1 score_2 score_3 score_4 score_4 13/ 30
41 skip2 skip1 while unprocessed hypotheses with no tied competitors remain do while the highest-scoring hypothesis has tied competitors do skip all tied competitors end while link the highest-scoring hypothesis discard all incompatible hypotheses end while a b e f g b h i score_1 score_3 score_4 score_4 13/ 30
42 skip2 while unprocessed hypotheses with no tied competitors remain do if the highest-scoring hypothesis has tied competitors do mark the constituents of all tied competitors end if while the highest-scoring hypothesis has a marked constituent do skip end while link the highest-scoring hypothesis a b score_1 discard all incompatible hypotheses a c score_1 unmark all constituents d c score_2 end while e f g b h i score_3 score_4 score_4 14/ 30
43 skip2 while unprocessed hypotheses with no tied competitors remain do if the highest-scoring hypothesis has tied competitors do mark the constituents of all tied competitors end if while the highest-scoring hypothesis has a marked constituent do skip end while link the highest-scoring hypothesis a b score_1 discard all incompatible hypotheses a c score_1 unmark all constituents d c score_2 end while e f g b h i score_3 score_4 score_4 14/ 30
44 skip2 while unprocessed hypotheses with no tied competitors remain do if the highest-scoring hypothesis has tied competitors do mark the constituents of all tied competitors end if while the highest-scoring hypothesis has a marked constituent do skip end while link the highest-scoring hypothesis a b score_1 discard all incompatible hypotheses a c score_1 unmark all constituents d c score_2 end while e f g b h i score_3 score_4 score_4 14/ 30
45 skip2 while unprocessed hypotheses with no tied competitors remain do if the highest-scoring hypothesis has tied competitors do mark the constituents of all tied competitors end if while the highest-scoring hypothesis has a marked constituent do skip end while link the highest-scoring hypothesis a b score_1 discard all incompatible hypotheses a c score_1 unmark all constituents d c score_2 end while a b is incompatible with e f e f g b h i score_3 score_4 score_4 14/ 30
46 skip2 while unprocessed hypotheses with no tied competitors remain do if the highest-scoring hypothesis has tied competitors do mark the constituents of all tied competitors end if while the highest-scoring hypothesis has a marked constituent do skip end while link the highest-scoring hypothesis discard all incompatible hypotheses unmark all constituents a c score_1 end while d c g b h i score_2 score_4 score_4 14/ 30
47 4an1 15/ 30
48 4an1 High"scoring improper lexical links prevent the produ#ion of good non"lexical links Split the set of nonzero links into two sets: lexical links non"lexical links 15/ 30
49 4an1 High"scoring improper lexical links prevent the produ#ion BCof good non"lexical links A a AB B Split the set of nonzero links into two sets: b lexical links C c non"lexical links W w A$ $ W B$ $ X B$ $ Z C$ $ Y BC$ $ XY BC$ $ Z BC$ $ WZ AB$ $ WZ 15/ 30 WZ XY X x Y y Z z
50 4an1 AB WZ High"scoring improper lexical links prevent the produ#ion A BCof good W non"lexical XY links Z a B C w z Split the set of nonzero links X into Ytwo sets: b c x y lexical links lexical non'lexical non"lexical links A$ $ W B$ $ X B$ $ Z C$ $ Y BC$ $ Z BC$ $ XY BC$ $ WZ AB$ $ WZ 15/ 30
51 4an1 High"scoring improper lexical links prevent the produ#ion of good non"lexical links Split the set of nonzero links into two sets: lexical links non"lexical links Perform sele#ion fir among the non"lexical links then among the lexical links 15/ 30
52 5u$'&earch &ele3io" 16/ 30
53 5u$'&earch &ele3io" Backtracking Recursive Algorithm enumerate all possible combinations of non"crossing links ore all maximal combinations of non"crossing links The be combination of links is the highe"scoring maximal combination of links Ambiguity again 16/ 30
54 &tring'to'&tring (lignme6 17/ 30
55 &tring'to'&tring (lignme6 The sub"tree aligner operates on parsed data For many languages no parsers are available Retraining exiing parsers for new languages may require significant resources The ring"to"ring aligner operates on plain sentences 17/ 30
56 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links Sele# the be set of links as before Two links are incompatible if they contain nodes that are part of incompatible trees 18/ 30
57 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links X 1 Sele# the be set of links as before X X 4 Two links are incompatible if they contain X f nodes that are part of incompatible trees X a X 3 b X 2 X c X e X 18/ 30
58 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links X 1 Sele# the be set of links as before X X 4 Two links are incompatible if they contain X f nodes that are part of incompatible trees X a X 3 b X 2 X c X e X 18/ 30
59 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links X 1 Sele# the be set of links as before X X 4 Two links are incompatible if they contain X f nodes that are part of incompatible trees X a X 3 b X 2 X c X e X 18/ 30
60 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links X 1 Sele# the be set of links as before X X 4 Two links are incompatible if they contain X f nodes that are part of incompatible trees X a X 3 b X 2 X c X e X 18/ 30
61 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links X 1 Sele# the be set of links as before X X 4 Two links are incompatible if they contain X f nodes that are part of incompatible trees X a X 3 b X 2 X c X e X 18/ 30
62 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links X 1 Sele# the be set of links as before X Two links are incompatible if they contain X f nodes that are part of incompatible trees X a b X 4 X c X e X 18/ 30
63 &tring'to'&tring (lignme6 algorith7 Generate all possible binary trees for each sentence in the sentence pair Calculate scores for all possible links Sele# the be set of links as before Two links are incompatible if they contain nodes that are part of incompatible trees Only output linked nodes and nodes needed for ru#ural integrity 18/ 30
64 &tring'to'&tring (lignme6 Tree"to"ring and ring"to"tree modules used when a parser exis for one of the languages in queion based on the ring"to"ring module preserve original parse ru#ures 19/ 30
65 8e'&coring (lgorith7 20/ 30
66 8e'&coring (lgorith7 Use already fixed links to eablish the context for word alignments 20/ 30
67 8e'&coring (lgorith7 a c w z x y b 20/ 30
68 8e'&coring (lgorith7 a c w z x y b 20/ 30
69 8e'&coring (lgorith7 a c w z x y b 20/ 30
70 8e'&coring (lgorith7 a c w z x y b 20/ 30
71 8e'&coring (lgorith7 a b,, c w z x y 20/ 30
72 (lgorithm 9omplexity 21/ 30
73 (lgorithm 9omplexity Tree"to"Tree Alignment space: O%n& time: O%m 2 & String"to"String Alignment space: O%n 2 & time: O%m 4 & Full"Search Algorithm related to the TSP, albeit highly reri#ed Score calculation may be executed in parallel 21/ 30
74 9o/e/s of the :-tributio" 23/ 30
75 9o/e/s of the :-tributio" Available at an RSS feed is available for update notifications GPL licence README file C++ code of the sub"tree aligner Configuration and compilation scripts 23/ 30
76 8equired &o;war+ 24/ 30
77 8equired &o;war+ GCC 4.0+ GCC 4.2+ required for the compilation of parallel code boo for ring"to"ring, tree"to"ring and ring"to"tree modules Should compile on any UNIX syem 24/ 30
78 (venu. for Improveme6 26/ 30
79 (venu. for Improveme6 Add new scoring algorithms %eg. maximum"entropy"based& Research ways to reduce the number of scores that need to be calculated Store word"alignment data in memory in a more optimal way Research ways to integrate POS"tag data into the ru#ures generated by the ring"based modules Make the main features sele#able at run time, rather than at compile time 26/ 30
80 9oncl<io1 28/ 30
81 9oncl<io1 Developed a novel platform for the fa and robu generation of parallel treebanks The aligner can handle very large amounts of data There are many underresourced languages String"to"ring, ring"to"tree and tree"to"ring algorithms have been developed Please send your bug reports to bugs@ventsislavzhechev.eu 28/ 30
82 %hank you
Syntax-Based Decoding
Syntax-Based Decoding Philipp Koehn 9 November 2017 1 syntax-based models Synchronous Context Free Grammar Rules 2 Nonterminal rules NP DET 1 2 JJ 3 DET 1 JJ 3 2 Terminal rules N maison house NP la maison
Syntax-based Statistical Machine Translation
Syntax-based Statistical Machine Translation Philip Williams and Philipp Koehn 29 October 2014 Part I Part II - Introduction - Rule Extraction Part III - Decoding Part IV - Extensions Syntax-based Statistical
National Centre for Language Technology School of Computing Dublin City University
with with National Centre for Language Technology School of Computing Dublin City University Parallel treebanks A parallel treebank comprises: sentence pairs parsed word-aligned tree-aligned (Volk & Samuelsson,
Probabilistic Context-free Grammars
Probabilistic Context-free Grammars Computational Linguistics Alexander Koller 24 November 2017 The CKY Recognizer S NP VP NP Det N VP V NP V ate NP John Det a N sandwich i = 1 2 3 4 k = 2 3 4 5 S NP John
Part I - Introduction Part II - Rule Extraction Part III - Decoding Part IV - Extensions
Syntax-based Statistical Machine Translation Philip Williams and Philipp Koehn 29 October 2014 Part I - Introduction Part II - Rule Extraction Part III - Decoding Part IV - Extensions Syntax-based Statistical
LECTURER: BURCU CAN Spring
LECTURER: BURCU CAN 2017-2018 Spring Regular Language Hidden Markov Model (HMM) Context Free Language Context Sensitive Language Probabilistic Context Free Grammar (PCFG) Unrestricted Language PCFGs can
Probabilistic Context Free Grammars. Many slides from Michael Collins and Chris Manning
Probabilistic Context Free Grammars Many slides from Michael Collins and Chris Manning Overview I Probabilistic Context-Free Grammars (PCFGs) I The CKY Algorithm for parsing with PCFGs A Probabilistic
Probabilistic Context Free Grammars. Many slides from Michael Collins
Probabilistic Context Free Grammars Many slides from Michael Collins Overview I Probabilistic Context-Free Grammars (PCFGs) I The CKY Algorithm for parsing with PCFGs A Probabilistic Context-Free Grammar
A DOP Model for LFG. Rens Bod and Ronald Kaplan. Kathrin Spreyer Data-Oriented Parsing, 14 June 2005
A DOP Model for LFG Rens Bod and Ronald Kaplan Kathrin Spreyer Data-Oriented Parsing, 14 June 2005 Lexical-Functional Grammar (LFG) Levels of linguistic knowledge represented formally differently (non-monostratal):
Probabilistic Context-Free Grammars. Michael Collins, Columbia University
Probabilistic Context-Free Grammars Michael Collins, Columbia University Overview Probabilistic Context-Free Grammars (PCFGs) The CKY Algorithm for parsing with PCFGs A Probabilistic Context-Free Grammar
CKY & Earley Parsing. Ling 571 Deep Processing Techniques for NLP January 13, 2016
CKY & Earley Parsing Ling 571 Deep Processing Techniques for NLP January 13, 2016 No Class Monday: Martin Luther King Jr. Day CKY Parsing: Finish the parse Recognizer à Parser Roadmap Earley parsing Motivation:
Decoding and Inference with Syntactic Translation Models
Decoding and Inference with Syntactic Translation Models March 5, 2013 CFGs S NP VP VP NP V V NP NP CFGs S NP VP S VP NP V V NP NP CFGs S NP VP S VP NP V NP VP V NP NP CFGs S NP VP S VP NP V NP VP V NP
Advanced Natural Language Processing Syntactic Parsing
Advanced Natural Language Processing Syntactic Parsing Alicia Ageno ageno@cs.upc.edu Universitat Politècnica de Catalunya NLP statistical parsing 1 Parsing Review Statistical Parsing SCFG Inside Algorithm
Statistical Methods for NLP
Statistical Methods for NLP Stochastic Grammars Joakim Nivre Uppsala University Department of Linguistics and Philology joakim.nivre@lingfil.uu.se Statistical Methods for NLP 1(22) Structured Classification
Processing/Speech, NLP and the Web
CS460/626 : Natural Language Processing/Speech, NLP and the Web (Lecture 25 Probabilistic Parsing) Pushpak Bhattacharyya CSE Dept., IIT Bombay 14 th March, 2011 Bracketed Structure: Treebank Corpus [ S1[
Parsing with Context-Free Grammars
Parsing with Context-Free Grammars Berlin Chen 2005 References: 1. Natural Language Understanding, chapter 3 (3.1~3.4, 3.6) 2. Speech and Language Processing, chapters 9, 10 NLP-Berlin Chen 1 Grammars
Statistical Machine Translation
Statistical Machine Translation -tree-based models (cont.)- Artem Sokolov Computerlinguistik Universität Heidelberg Sommersemester 2015 material from P. Koehn, S. Riezler, D. Altshuler Bottom-Up Decoding
Parsing with Context-Free Grammars
Parsing with Context-Free Grammars CS 585, Fall 2017 Introduction to Natural Language Processing http://people.cs.umass.edu/~brenocon/inlp2017 Brendan O Connor College of Information and Computer Sciences
Parsing. Based on presentations from Chris Manning s course on Statistical Parsing (Stanford)
Parsing Based on presentations from Chris Manning s course on Statistical Parsing (Stanford) S N VP V NP D N John hit the ball Levels of analysis Level Morphology/Lexical POS (morpho-synactic), WSD Elements
Natural Language Processing : Probabilistic Context Free Grammars. Updated 5/09
Natural Language Processing : Probabilistic Context Free Grammars Updated 5/09 Motivation N-gram models and HMM Tagging only allowed us to process sentences linearly. However, even simple sentences require
Parsing with CFGs L445 / L545 / B659. Dept. of Linguistics, Indiana University Spring Parsing with CFGs. Direction of processing
L445 / L545 / B659 Dept. of Linguistics, Indiana University Spring 2016 1 / 46 : Overview Input: a string Output: a (single) parse tree A useful step in the process of obtaining meaning We can view the
Parsing with CFGs. Direction of processing. Top-down. Bottom-up. Left-corner parsing. Chart parsing CYK. Earley 1 / 46.
: Overview L545 Dept. of Linguistics, Indiana University Spring 2013 Input: a string Output: a (single) parse tree A useful step in the process of obtaining meaning We can view the problem as searching
Review. Earley Algorithm Chapter Left Recursion. Left-Recursion. Rule Ordering. Rule Ordering
Review Earley Algorithm Chapter 13.4 Lecture #9 October 2009 Top-Down vs. Bottom-Up Parsers Both generate too many useless trees Combine the two to avoid over-generation: Top-Down Parsing with Bottom-Up
A Context-Free Grammar
Statistical Parsing A Context-Free Grammar S VP VP Vi VP Vt VP VP PP DT NN PP PP P Vi sleeps Vt saw NN man NN dog NN telescope DT the IN with IN in Ambiguity A sentence of reasonable length can easily
Random Generation of Nondeterministic Tree Automata
Random Generation of Nondeterministic Tree Automata Thomas Hanneforth 1 and Andreas Maletti 2 and Daniel Quernheim 2 1 Department of Linguistics University of Potsdam, Germany 2 Institute for Natural Language
Spectral Unsupervised Parsing with Additive Tree Metrics
Spectral Unsupervised Parsing with Additive Tree Metrics Ankur Parikh, Shay Cohen, Eric P. Xing Carnegie Mellon, University of Edinburgh Ankur Parikh 2014 1 Overview Model: We present a novel approach
Introduction to Theoretical Computer Science. Motivation. Automata = abstract computing devices
Introduction to Theoretical Computer Science Motivation Automata = abstract computing devices Turing studied Turing Machines (= computers) before there were any real computers We will also look at simpler
Lecture 5: UDOP, Dependency Grammars
Lecture 5: UDOP, Dependency Grammars Jelle Zuidema ILLC, Universiteit van Amsterdam Unsupervised Language Learning, 2014 Generative Model objective PCFG PTSG CCM DMV heuristic Wolff (1984) UDOP ML IO K&M
A Syntax-based Statistical Machine Translation Model. Alexander Friedl, Georg Teichtmeister
A Syntax-based Statistical Machine Translation Model Alexander Friedl, Georg Teichtmeister 4.12.2006 Introduction The model Experiment Conclusion Statistical Translation Model (STM): - mathematical model
S NP VP 0.9 S VP 0.1 VP V NP 0.5 VP V 0.1 VP V PP 0.1 NP NP NP 0.1 NP NP PP 0.2 NP N 0.7 PP P NP 1.0 VP NP PP 1.0. N people 0.
/6/7 CS 6/CS: Natural Language Processing Instructor: Prof. Lu Wang College of Computer and Information Science Northeastern University Webpage: www.ccs.neu.edu/home/luwang The grammar: Binary, no epsilons,.9..5
Statistical Machine Translation. Part III: Search Problem. Complexity issues. DP beam-search: with single and multi-stacks
Statistical Machine Translation Marcello Federico FBK-irst Trento, Italy Galileo Galilei PhD School - University of Pisa Pisa, 7-19 May 008 Part III: Search Problem 1 Complexity issues A search: with single
Penn Treebank Parsing. Advanced Topics in Language Processing Stephen Clark
Penn Treebank Parsing Advanced Topics in Language Processing Stephen Clark 1 The Penn Treebank 40,000 sentences of WSJ newspaper text annotated with phrasestructure trees The trees contain some predicate-argument
Context-Free Parsing: CKY & Earley Algorithms and Probabilistic Parsing
Context-Free Parsing: CKY & Earley Algorithms and Probabilistic Parsing Natural Language Processing CS 4120/6120 Spring 2017 Northeastern University David Smith with some slides from Jason Eisner & Andrew
AN ABSTRACT OF THE DISSERTATION OF
AN ABSTRACT OF THE DISSERTATION OF Kai Zhao for the degree of Doctor of Philosophy in Computer Science presented on May 30, 2017. Title: Structured Learning with Latent Variables: Theory and Algorithms
Analysing Soft Syntax Features and Heuristics for Hierarchical Phrase Based Machine Translation
Analysing Soft Syntax Features and Heuristics for Hierarchical Phrase Based Machine Translation David Vilar, Daniel Stein, Hermann Ney IWSLT 2008, Honolulu, Hawaii 20. October 2008 Human Language Technology
Algorithms for Syntax-Aware Statistical Machine Translation
Algorithms for Syntax-Aware Statistical Machine Translation I. Dan Melamed, Wei Wang and Ben Wellington ew York University Syntax-Aware Statistical MT Statistical involves machine learning (ML) seems crucial
Recap: Lexicalized PCFGs (Fall 2007): Lecture 5 Parsing and Syntax III. Recap: Charniak s Model. Recap: Adding Head Words/Tags to Trees
Recap: Lexicalized PCFGs We now need to estimate rule probabilities such as P rob(s(questioned,vt) NP(lawyer,NN) VP(questioned,Vt) S(questioned,Vt)) 6.864 (Fall 2007): Lecture 5 Parsing and Syntax III
A Supertag-Context Model for Weakly-Supervised CCG Parser Learning
A Supertag-Context Model for Weakly-Supervised CCG Parser Learning Dan Garrette Chris Dyer Jason Baldridge Noah A. Smith U. Washington CMU UT-Austin CMU Contributions 1. A new generative model for learning
Aspects of Tree-Based Statistical Machine Translation
Aspects of Tree-Based Statistical Machine Translation Marcello Federico Human Language Technology FBK 2014 Outline Tree-based translation models: Synchronous context free grammars Hierarchical phrase-based
Improving Neural Parsing by Disentangling Model Combination and Reranking Effects. Daniel Fried*, Mitchell Stern* and Dan Klein UC Berkeley
Improving Neural Parsing by Disentangling Model Combination and Reranking Effects Daniel Fried*, Mitchell Stern* and Dan Klein UC erkeley Top-down generative models Top-down generative models S VP The
Quasi-Synchronous Phrase Dependency Grammars for Machine Translation. lti
Quasi-Synchronous Phrase Dependency Grammars for Machine Translation Kevin Gimpel Noah A. Smith 1 Introduction MT using dependency grammars on phrases Phrases capture local reordering and idiomatic translations
Aspects of Tree-Based Statistical Machine Translation
Aspects of Tree-Based tatistical Machine Translation Marcello Federico (based on slides by Gabriele Musillo) Human Language Technology FBK-irst 2011 Outline Tree-based translation models: ynchronous context
Multilevel Coarse-to-Fine PCFG Parsing
Multilevel Coarse-to-Fine PCFG Parsing Eugene Charniak, Mark Johnson, Micha Elsner, Joseph Austerweil, David Ellis, Isaac Haxton, Catherine Hill, Shrivaths Iyengar, Jeremy Moore, Michael Pozar, and Theresa
CS 545 Lecture XVI: Parsing
CS 545 Lecture XVI: Parsing brownies_choco81@yahoo.com brownies_choco81@yahoo.com Benjamin Snyder Parsing Given a grammar G and a sentence x = (x1, x2,..., xn), find the best parse tree. We re not going
A Probabilistic Forest-to-String Model for Language Generation from Typed Lambda Calculus Expressions
A Probabilistic Forest-to-String Model for Language Generation from Typed Lambda Calculus Expressions Wei Lu and Hwee Tou Ng National University of Singapore 1/26 The Task (Logical Form) λx 0.state(x 0
Lecture 7: Introduction to syntax-based MT
Lecture 7: Introduction to syntax-based MT Andreas Maletti Statistical Machine Translation Stuttgart December 16, 2011 SMT VII A. Maletti 1 Lecture 7 Goals Overview Tree substitution grammars (tree automata)
Context- Free Parsing with CKY. October 16, 2014
Context- Free Parsing with CKY October 16, 2014 Lecture Plan Parsing as Logical DeducBon Defining the CFG recognibon problem BoHom up vs. top down Quick review of Chomsky normal form The CKY algorithm
This kind of reordering is beyond the power of finite transducers, but a synchronous CFG can do this.
Chapter 12 Synchronous CFGs Synchronous context-free grammars are a generalization of CFGs that generate pairs of related strings instead of single strings. They are useful in many situations where one
Natural Language Processing CS Lecture 06. Razvan C. Bunescu School of Electrical Engineering and Computer Science
Natural Language Processing CS 6840 Lecture 06 Razvan C. Bunescu School of Electrical Engineering and Computer Science bunescu@ohio.edu Statistical Parsing Define a probabilistic model of syntax P(T S):
The Infinite PCFG using Hierarchical Dirichlet Processes
S NP VP NP PRP VP VBD NP NP DT NN PRP she VBD heard DT the NN noise S NP VP NP PRP VP VBD NP NP DT NN PRP she VBD heard DT the NN noise S NP VP NP PRP VP VBD NP NP DT NN PRP she VBD heard DT the NN noise
PCFGs 2 L645 / B659. Dept. of Linguistics, Indiana University Fall PCFGs 2. Questions. Calculating P(w 1m ) Inside Probabilities
1 / 22 Inside L645 / B659 Dept. of Linguistics, Indiana University Fall 2015 Inside- 2 / 22 for PCFGs 3 questions for Probabilistic Context Free Grammars (PCFGs): What is the probability of a sentence
Introduction to Computers & Programming
16.070 Introduction to Computers & Programming Theory of computation: What is a computer? FSM, Automata Prof. Kristina Lundqvist Dept. of Aero/Astro, MIT Models of Computation What is a computer? If you
Latent Variable Models in NLP
Latent Variable Models in NLP Aria Haghighi with Slav Petrov, John DeNero, and Dan Klein UC Berkeley, CS Division Latent Variable Models Latent Variable Models Latent Variable Models Observed Latent Variable
THEORY OF COMPILATION
Lecture 04 Syntax analysis: top-down and bottom-up parsing THEORY OF COMPILATION EranYahav 1 You are here Compiler txt Source Lexical Analysis Syntax Analysis Parsing Semantic Analysis Inter. Rep. (IR)
NLP Programming Tutorial 11 - The Structured Perceptron
NLP Programming Tutorial 11 - The Structured Perceptron Graham Neubig Nara Institute of Science and Technology (NAIST) 1 Prediction Problems Given x, A book review Oh, man I love this book! This book is
CFLs and Regular Languages. CFLs and Regular Languages. CFLs and Regular Languages. Will show that all Regular Languages are CFLs. Union.
We can show that every RL is also a CFL Since a regular grammar is certainly context free. We can also show by only using Regular Expressions and Context Free Grammars That is what we will do in this half.
A* Search. 1 Dijkstra Shortest Path
A* Search Consider the eight puzzle. There are eight tiles numbered 1 through 8 on a 3 by three grid with nine locations so that one location is left empty. We can move by sliding a tile adjacent to the
Features of Statistical Parsers
Features of tatistical Parsers Preliminary results Mark Johnson Brown University TTI, October 2003 Joint work with Michael Collins (MIT) upported by NF grants LI 9720368 and II0095940 1 Talk outline tatistical
A Translation from Attribute Grammars to CatamorpMsms
A Translation from Attribute Grammars to CatamorpMsms Maarten Fokkinga, Johan Jeuring, Lambert Meertens, Erik Meijer November 9, 1990 Let A G be an attribute grammar, with underlying context free grammar
IBM Model 1 for Machine Translation
IBM Model 1 for Machine Translation Micha Elsner March 28, 2014 2 Machine translation A key area of computational linguistics Bar-Hillel points out that human-like translation requires understanding of
Multiword Expression Identification with Tree Substitution Grammars
Multiword Expression Identification with Tree Substitution Grammars Spence Green, Marie-Catherine de Marneffe, John Bauer, and Christopher D. Manning Stanford University EMNLP 2011 Main Idea Use syntactic
Attendee information. Seven Lectures on Statistical Parsing. Phrase structure grammars = context-free grammars. Assessment.
even Lectures on tatistical Parsing Christopher Manning LA Linguistic Institute 7 LA Lecture Attendee information Please put on a piece of paper: ame: Affiliation: tatus (undergrad, grad, industry, prof,
Probabilistic Context-Free Grammar
Probabilistic Context-Free Grammar Petr Horáček, Eva Zámečníková and Ivana Burgetová Department of Information Systems Faculty of Information Technology Brno University of Technology Božetěchova 2, 612
Natural Language Processing
SFU NatLangLab Natural Language Processing Anoop Sarkar anoopsarkar.github.io/nlp-class Simon Fraser University September 27, 2018 0 Natural Language Processing Anoop Sarkar anoopsarkar.github.io/nlp-class
Parsing. Probabilistic CFG (PCFG) Laura Kallmeyer. Winter 2017/18. Heinrich-Heine-Universität Düsseldorf 1 / 22
Parsing Probabilistic CFG (PCFG) Laura Kallmeyer Heinrich-Heine-Universität Düsseldorf Winter 2017/18 1 / 22 Table of contents 1 Introduction 2 PCFG 3 Inside and outside probability 4 Parsing Jurafsky
Unit 2: Tree Models. CS 562: Empirical Methods in Natural Language Processing. Lectures 19-23: Context-Free Grammars and Parsing
CS 562: Empirical Methods in Natural Language Processing Unit 2: Tree Models Lectures 19-23: Context-Free Grammars and Parsing Oct-Nov 2009 Liang Huang (lhuang@isi.edu) Big Picture we have already covered...
The relation of surprisal and human processing
The relation of surprisal and human processing difficulty Information Theory Lecture Vera Demberg and Matt Crocker Information Theory Lecture, Universität des Saarlandes April 19th, 2015 Information theory
6.891: Lecture 24 (December 8th, 2003) Kernel Methods
6.891: Lecture 24 (December 8th, 2003) Kernel Methods Overview ffl Recap: global linear models ffl New representations from old representations ffl computational trick ffl Kernels for NLP structures ffl
Discriminative Training. March 4, 2014
Discriminative Training March 4, 2014 Noisy Channels Again p(e) source English Noisy Channels Again p(e) p(g e) source English German Noisy Channels Again p(e) p(g e) source English German decoder e =
Lecture 9: Decoding. Andreas Maletti. Stuttgart January 20, Statistical Machine Translation. SMT VIII A. Maletti 1
Lecture 9: Decoding Andreas Maletti Statistical Machine Translation Stuttgart January 20, 2012 SMT VIII A. Maletti 1 Lecture 9 Last time Synchronous grammars (tree transducers) Rule extraction Weight training
Computational Linguistics
Computational Linguistics Dependency-based Parsing Clayton Greenberg Stefan Thater FR 4.7 Allgemeine Linguistik (Computerlinguistik) Universität des Saarlandes Summer 2016 Acknowledgements These slides
Introduction to Computational Linguistics
Introduction to Computational Linguistics Olga Zamaraeva (2018) Based on Bender (prev. years) University of Washington May 3, 2018 1 / 101 Midterm Project Milestone 2: due Friday Assgnments 4& 5 due dates
Computational Linguistics. Acknowledgements. Phrase-Structure Trees. Dependency-based Parsing
Computational Linguistics Dependency-based Parsing Dietrich Klakow & Stefan Thater FR 4.7 Allgemeine Linguistik (Computerlinguistik) Universität des Saarlandes Summer 2013 Acknowledgements These slides
NLP Homework: Dependency Parsing with Feed-Forward Neural Network
NLP Homework: Dependency Parsing with Feed-Forward Neural Network Submission Deadline: Monday Dec. 11th, 5 pm 1 Background on Dependency Parsing Dependency trees are one of the main representations used
Constituency Parsing
CS5740: Natural Language Processing Spring 2017 Constituency Parsing Instructor: Yoav Artzi Slides adapted from Dan Klein, Dan Jurafsky, Chris Manning, Michael Collins, Luke Zettlemoyer, Yejin Choi, and
CS626: NLP, Speech and the Web. Pushpak Bhattacharyya CSE Dept., IIT Bombay Lecture 14: Parsing Algorithms 30 th August, 2012
CS626: NLP, Speech and the Web Pushpak Bhattacharyya CSE Dept., IIT Bombay Lecture 14: Parsing Algorithms 30 th August, 2012 Parsing Problem Semantics Part of Speech Tagging NLP Trinity Morph Analysis
Lab 12: Structured Prediction
December 4, 2014 Lecture plan structured perceptron application: confused messages application: dependency parsing structured SVM Class review: from modelization to classification What does learning mean?
Compositions of Bottom-Up Tree Series Transformations
Compositions of Bottom-Up Tree Series Transformations Andreas Maletti a Technische Universität Dresden Fakultät Informatik D 01062 Dresden, Germany maletti@tcs.inf.tu-dresden.de May 17, 2005 1. Motivation
13A. Computational Linguistics. 13A. Log-Likelihood Dependency Parsing. CSC 2501 / 485 Fall 2017
Computational Linguistics CSC 2501 / 485 Fall 2017 13A 13A. Log-Likelihood Dependency Parsing Gerald Penn Department of Computer Science, University of Toronto Based on slides by Yuji Matsumoto, Dragomir
An Efficient Context-Free Parsing Algorithm. Speakers: Morad Ankri Yaniv Elia
An Efficient Context-Free Parsing Algorithm Speakers: Morad Ankri Yaniv Elia Yaniv: Introduction Terminology Informal Explanation The Recognizer Morad: Example Time and Space Bounds Empirical results Practical
Syntax Analysis Part I. Position of a Parser in the Compiler Model. The Parser. Chapter 4
1 Syntax Analysis Part I Chapter 4 COP5621 Compiler Construction Copyright Robert van ngelen, Flora State University, 2007 Position of a Parser in the Compiler Model 2 Source Program Lexical Analyzer Lexical
CTL satisfability via tableau A C++ implementation
CTL satisfability via tableau A C++ implementation Nicola Prezza April 30, 2015 Outline 1 Introduction 2 Parsing 3 Data structures 4 Initial sets of labels and edges 5 Cull 6 Examples and benchmarks The
Bringing machine learning & compositional semantics together: central concepts
Bringing machine learning & compositional semantics together: central concepts https://githubcom/cgpotts/annualreview-complearning Chris Potts Stanford Linguistics CS 244U: Natural language understanding
Syntactical analysis. Syntactical analysis. Syntactical analysis. Syntactical analysis
Context-free grammars Derivations Parse Trees Left-recursive grammars Top-down parsing non-recursive predictive parsers construction of parse tables Bottom-up parsing shift/reduce parsers LR parsers GLR
CS460/626 : Natural Language
CS460/626 : Natural Language Processing/Speech, NLP and the Web (Lecture 23, 24 Parsing Algorithms; Parsing in case of Ambiguity; Probabilistic Parsing) Pushpak Bhattacharyya CSE Dept., IIT Bombay 8 th,
Computational Linguistics II: Parsing
Computational Linguistics II: Parsing Left-corner-Parsing Frank Richter & Jan-Philipp Söhn fr@sfs.uni-tuebingen.de, jp.soehn@uni-tuebingen.de January 15th, 2007 Richter/Söhn (WS 2006/07) Computational
Introduction to Bottom-Up Parsing
Introduction to Bottom-Up Parsing Outline Review LL parsing Shift-reduce parsing The LR parsing algorithm Constructing LR parsing tables Compiler Design 1 (2011) 2 Top-Down Parsing: Review Top-down parsing
SYLLABUS. Introduction to Finite Automata, Central Concepts of Automata Theory. CHAPTER - 3 : REGULAR EXPRESSIONS AND LANGUAGES
Contents i SYLLABUS UNIT - I CHAPTER - 1 : AUT UTOMA OMATA Introduction to Finite Automata, Central Concepts of Automata Theory. CHAPTER - 2 : FINITE AUT UTOMA OMATA An Informal Picture of Finite Automata,
EXAM. CS331 Compiler Design Spring Please read all instructions, including these, carefully
EXAM Please read all instructions, including these, carefully There are 7 questions on the exam, with multiple parts. You have 3 hours to work on the exam. The exam is open book, open notes. Please write
{Probabilistic Stochastic} Context-Free Grammars (PCFGs)
{Probabilistic Stochastic} Context-Free Grammars (PCFGs) 116 The velocity of the seismic waves rises to... S NP sg VP sg DT NN PP risesto... The velocity IN NP pl of the seismic waves 117 PCFGs APCFGGconsists
X-bar theory. X-bar :
is one of the greatest contributions of generative school in the filed of knowledge system. Besides linguistics, computer science is greatly indebted to Chomsky to have propounded the theory of x-bar.
Natural Language Processing 1. lecture 7: constituent parsing. Ivan Titov. Institute for Logic, Language and Computation
atural Language Processing 1 lecture 7: constituent parsing Ivan Titov Institute for Logic, Language and Computation Outline Syntax: intro, CFGs, PCFGs PCFGs: Estimation CFGs: Parsing PCFGs: Parsing Parsing
CS 6120/CS4120: Natural Language Processing
CS 6120/CS4120: Natural Language Processing Instructor: Prof. Lu Wang College of Computer and Information Science Northeastern University Webpage: www.ccs.neu.edu/home/luwang Assignment/report submission
Universität Potsdam Institut für Informatik Lehrstuhl Maschinelles Lernen. Machine Translation. Uwe Dick
Universität Potsdam Institut für Informatik Lehrstuhl Maschinelles Lernen Machine Translation Uwe Dick Google Translate Rosetta Stone Hieroglyphs Demotic Greek Machine Translation Automatically translate
Applications of Tree Automata Theory Lecture VI: Back to Machine Translation
Applications of Tree Automata Theory Lecture VI: Back to Machine Translation Andreas Maletti Institute of Computer Science Universität Leipzig, Germany on leave from: Institute for Natural Language Processing
Discriminative Training
Discriminative Training February 19, 2013 Noisy Channels Again p(e) source English Noisy Channels Again p(e) p(g e) source English German Noisy Channels Again p(e) p(g e) source English German decoder
Efficient discovery of the top-k optimal dependency rules with Fisher s exact test of significance
Efficient discovery of the top-k optimal dependency rules with Fisher s exact test of significance Wilhelmiina Hämäläinen epartment of omputer Science University of Helsinki Finland whamalai@cs.helsinki.fi
Introduction to Bottom-Up Parsing
Outline Introduction to Bottom-Up Parsing Review LL parsing Shift-reduce parsing he LR parsing algorithm Constructing LR parsing tables Compiler Design 1 (2011) 2 op-down Parsing: Review op-down parsing
Introduction to Automata
Introduction to Automata Seungjin Choi Department of Computer Science and Engineering Pohang University of Science and Technology 77 Cheongam-ro, Nam-gu, Pohang 37673, Korea seungjin@postech.ac.kr 1 /
Natural Language Processing. Lecture 13: More on CFG Parsing
Natural Language Processing Lecture 13: More on CFG Parsing Probabilistc/Weighted Parsing Example: ambiguous parse Probabilistc CFG Ambiguous parse w/probabilites 0.05 0.05 0.20 0.10 0.30 0.20 0.60 0.75