Normalized alignment of dependency trees for detecting textual entailment

16
Normalized alignment of dependency trees for detecting textual entailment Erwin Marsi & Emiel Krahmer Tilburg University Wauter Bosma & Mariët Theune University of Twente

description

Normalized alignment of dependency trees for detecting textual entailment. Erwin Marsi & Emiel Krahmer Tilburg University. Wauter Bosma & Mariët Theune University of Twente. Basic idea. A true hypothesis is included in the text, allowing omission and rephrasing - PowerPoint PPT Presentation

Transcript of Normalized alignment of dependency trees for detecting textual entailment

Page 1: Normalized alignment of dependency trees for detecting textual entailment

Normalized alignment of dependency trees for detecting textual entailment

Erwin Marsi & Emiel Krahmer

Tilburg University

Wauter Bosma & Mariët Theune

University of Twente

Page 2: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 2

Basic idea

• A true hypothesis is included in the text, allowing omission and rephrasing

Text: The Rolling Stones kicked off their latest tour on Sunday with a concert at Boston's Fenway Park.

Hypothesis: The Rolling Stones have begun their latest tour with a concert in Boston.

Entailment: True

• Omissions:– on Sunday– Fenway Park

• Paraphrases:– kicked off begun– Boston's Fenway Park Boston

Page 3: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 3

Matching surface words alone is not sufficient...

• Variation in surface realization perfect word match is no guarantee for entailment

• Using syntactic analysis– for syntactic normalization

– to match on hierarchical relations among constituents

Example: “He became a boxing referee in 1964, and became well-known […]” “He became well-known in 1964”

Page 4: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 4

Preprocessing

• Input: T-H pairs in XML• Processing pipeline:

1. Sentence splitting, MXTERMINATOR (Reynar & Ratnaparkhi, 1997)

2. Tokenization, Penn Treebank SED script

3. POS tagging with PTB POS tags using Mbt (van den Bosch et al)

4. Lemmatizing using Memory-based learning (van den Bosch et al)

5. Dependency parsing using Maltparser trained on PTB (Nivre & Scholz, 2004)

6. Syntactic normalization

• Output: T-H dependency tree(s) pairs in XML

Page 5: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 5

Syntactic Normalization

• Three types of syntactic normalization:– Auxiliary reduction

– Passive to active form

– Copula reduction

Page 6: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 6

Auxiliary Reduction

• Auxiliaries of progressive and perfective tense are removed• Their children are attached to the remaining content verb• The same goes for modal verbs, and for do in the do-support

function.

Example: “demand for ivory has dropped” “demand for ivory dropped”

Example: “legalization does not solve any social problems” “legalization not solves any social problems”

Page 7: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 7

Passive to Active Form

• The passive form auxiliary is removed• The original subject becomes object• Where possible, a by-phrase becomes the subject

Example: “Ahmedinejad was attacked by the US” “the US attacked Ahmedinejad”

Page 8: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 8

Copula Reduction

• Copular verbs are removed by attaching the predicate as a daughter to the subject

Example: “Microsoft Corp. is a partner of Intel Corp.” “Microsoft Corp., a partner of Intel Corp.”

Page 9: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 9

Alignment of Dependency Trees

• Tree alignment algorithm based on (Meyers, Yangarbar and Grishman, 1996)

• Searches for an optimal alignment of the nodes of the text tree to the nodes of the hypothesis tree

• Tree alignment is a function of:1. how well the words of the two nodes match

2. recursively, the weighted alignment score for each of the aligned daughter nodes

Page 10: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 10

Word Matching

• function WordMatch(wt,wh) -> [0,1] maps text-hypothesis word pairs to a similarity score

• returns 1 if – wt is identical to wh

– the lemma of wt is identical to the lemma of wh

– wt is a synonym of wh (lookup in EuroWordnet with lemma & POS)

– wh is a hypernym of wt (idem)

• returns similarity from automatically derived thesaurus if > 0.1 (Lin’s dependency-based thesaurus)

• otherwise returns 0• also match on phrasal verbs

– e.g. “kick off“ is a synonym of “begin“

Page 11: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 11

Alignment example

Text: The development of agriculture by early humans, roughly 10,000 years ago, was also harmful to many natural ecosystems as they were systematically destroyed and replaced with artificial versions.

Hypothesis: Humans existed 10,000 years ago.Entailment: True

Page 12: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 12

Alignment example (cont’d)

Page 13: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 13

Entailment prediction

• Prediction rule:

IF top node of the hypothesis is aligned AND score > threshold

THEN

entailment = true

ELSE

entailment = false

• Threshold and parameters of tree alignment algorithm (skip penalty) optimized per task

Page 14: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 14

Results

Task DevO DevS Test

IE 56.0 53.0 52.0

IR 61.0 58.0 58.5

QA 60.0 57.5 62.5

SUM 72.0 72.0 69.0

Overall 62.25 60.1 60.5

Percentage entailment accuracy (n=800)

Page 15: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 15

Problems

• Many parses contain errors due to syntactic ambiguity and propagation of– Spelling errors

– Tokenization errors

– POS errors

– broken dependency trees

• Consequently, syntactic normalization & alignment failed• Dependency relations did not help

Page 16: Normalized alignment of dependency trees for detecting textual entailment

April 10 2006 RTE2 Workshop 16

Discussion & Conclusion

• There are many forms of textual entailment that we cannot recognize automatically...– Paraphrasing

– Co-reference resolution

– Ellipsis

– Condition/modality

– Inference

– Common sense / world knowledge

• RTE requires a combination of deep NLP, common sense knowledge and reasoning