Evaluating Word Order Recursively over Permutation-Forests Milo s - - PowerPoint PPT Presentation

evaluating word order recursively over permutation forests
SMART_READER_LITE
LIVE PREVIEW

Evaluating Word Order Recursively over Permutation-Forests Milo s - - PowerPoint PPT Presentation

Evaluating Word Order Recursively over Permutation-Forests Milo s Stanojevi c and Khalil Simaan October 25, 2014 What is wrong with existing metrics related to word order BLEU and many others have local view of word order (n-gram


slide-1
SLIDE 1

Evaluating Word Order Recursively over Permutation-Forests

Miloˇ s Stanojevi´ c and Khalil Sima’an October 25, 2014

slide-2
SLIDE 2

What is wrong with existing metrics related to word order

◮ BLEU and many others have local view of word order

(n-gram window) which is not good for long distance reordering.

slide-3
SLIDE 3

What is wrong with existing metrics related to word order

◮ BLEU and many others have local view of word order

(n-gram window) which is not good for long distance reordering.

◮ We need better representation that would allow a global

view – permutations (LRscore, RIBES, FuzzyScore)

slide-4
SLIDE 4

What is wrong with existing metrics related to word order

◮ BLEU and many others have local view of word order

(n-gram window) which is not good for long distance reordering.

◮ We need better representation that would allow a global

view – permutations (LRscore, RIBES, FuzzyScore)

◮ Problem: not hierarchical and not flexible

slide-5
SLIDE 5

What is wrong with existing metrics related to word order

◮ BLEU and many others have local view of word order

(n-gram window) which is not good for long distance reordering.

◮ We need better representation that would allow a global

view – permutations (LRscore, RIBES, FuzzyScore)

◮ Problem: not hierarchical and not flexible ◮ Permutation Trees (PETs) might come handy

slide-6
SLIDE 6

What is wrong with existing metrics related to word order

◮ BLEU and many others have local view of word order

(n-gram window) which is not good for long distance reordering.

◮ We need better representation that would allow a global

view – permutations (LRscore, RIBES, FuzzyScore)

◮ Problem: not hierarchical and not flexible ◮ Permutation Trees (PETs) might come handy ◮ Our metric computes its score in a way similar to PCFG on

these hierarchical structures

slide-7
SLIDE 7

Recursive metrics

slide-8
SLIDE 8

Recursive metrics

slide-9
SLIDE 9

Recursive metrics

slide-10
SLIDE 10

Recursive metrics

slide-11
SLIDE 11

Recursive metrics

slide-12
SLIDE 12

Recursive metrics

slide-13
SLIDE 13

Recursive metrics

slide-14
SLIDE 14

PETscore(·) and PEFscore(·)

PETscore(node) =β opScore(node.op)+ (1 − β)

  • c∈node.children

PETscore(c)

slide-15
SLIDE 15

PETscore(·) and PEFscore(·)

PETscore(node) =β opScore(node.op)+ (1 − β)

  • c∈node.children

PETscore(c) But, there might be (exponentially) many PETs for a single permutation! PEFscore(π) =

  • t∈PEF(π) PETscore(t)

#PETs Can be efficiently computed with a version of Inside algorithm.

slide-16
SLIDE 16

Results into English (scaled Kendall sent level)

slide-17
SLIDE 17

Results out of English (scaled Kendall sent level)

slide-18
SLIDE 18

Conclusion

◮ Consider all factorizations (PEF)

slide-19
SLIDE 19

Conclusion

◮ Consider all factorizations (PEF) ◮ Do it hierarchically

slide-20
SLIDE 20

Conclusion

◮ Consider all factorizations (PEF) ◮ Do it hierarchically ◮ Metric is available online together with BEER

slide-21
SLIDE 21

Conclusion

◮ Consider all factorizations (PEF) ◮ Do it hierarchically ◮ Metric is available online together with BEER ◮ Come to see the poster

slide-22
SLIDE 22

Conclusion

◮ Consider all factorizations (PEF) ◮ Do it hierarchically ◮ Metric is available online together with BEER ◮ Come to see the poster ◮ Thank you