Difference between revisions of "2009 NaturalLanguageInference"

From GM-RKB
Jump to: navigation, search
(Cited By)
(ContinuousReplacement)
(Tag: continuous replacement)
Line 1: Line 1:
* ([[2009_NaturalLanguageInference|Maccartney, 2009]]) ⇒ [[author::Bill Maccartney]]. ([[year::2009]]). “[http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.156.2685&rep=rep1&type=pdf Natural Language Inference].” Stanford University. ISBN:978-1-109-24088-7  
+
* ([[2009_NaturalLanguageInference|Maccartney, 2009]]) [[author::Bill Maccartney]]. ([[year::2009]]). “[http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.156.2685&rep=rep1&type=pdf Natural Language Inference].” Stanford University. ISBN:978-1-109-24088-7  
  
 
<B>Subject Headings:</B>  
 
<B>Subject Headings:</B>  
  
==Notes==
+
== Notes ==
  
==Cited By==
+
== Cited By ==
 
* [[Google Scholar]]: ~ 147 [http://scholar.google.com/scholar?q=%222009%22+Natural+Language+Inference Citations]
 
* [[Google Scholar]]: ~ 147 [http://scholar.google.com/scholar?q=%222009%22+Natural+Language+Inference Citations]
 
* [[DL-ACM]]: ~ 13 [http://dl.acm.org/citation.cfm?id=1751277&preflayout=flat#citedby Citations]
 
* [[DL-ACM]]: ~ 13 [http://dl.acm.org/citation.cfm?id=1751277&preflayout=flat#citedby Citations]
 
* [[Semantic Scholar]] ~ 105 [https://www.semanticscholar.org/paper/Natural-language-inference-Manning-MacCartney/8314f8eef3b64054bfc00607507a92de92fb7c85#citing-papers Citations]
 
* [[Semantic Scholar]] ~ 105 [https://www.semanticscholar.org/paper/Natural-language-inference-Manning-MacCartney/8314f8eef3b64054bfc00607507a92de92fb7c85#citing-papers Citations]
  
==Quotes==
+
== Quotes ==
  
  
===Abstract===
+
=== Abstract ===
  
 
[[Inference]] has been a central [[topic in artificial intelligence]] from the [[start]], but while [[automatic method]]s for [[formal deduction]] have [[advanced tremendously]], comparatively little [[progress]] has been made on the [[problem of <italic>natural language inference</italic> (NLI)]], that is, determining whether a [[natural language hypothesi]]s <italic>h</italic> can justifiably be [[inferred]] from a [[natural language premise]] <italic>p.</italic> The challenges of [[NLI]] are quite different from those encountered in [[formal deduction]]: the emphasis is on [[informal reasoning]], [[lexical semantic knowledge]], and [[variability]] of [[linguistic expression.This dissertation]] explores a range of [[approaches to NLI]], beginning with [[method]]s which are robust but [[approximate]], and proceeding to progressively more precise [[approach]]es. </s>
 
[[Inference]] has been a central [[topic in artificial intelligence]] from the [[start]], but while [[automatic method]]s for [[formal deduction]] have [[advanced tremendously]], comparatively little [[progress]] has been made on the [[problem of <italic>natural language inference</italic> (NLI)]], that is, determining whether a [[natural language hypothesi]]s <italic>h</italic> can justifiably be [[inferred]] from a [[natural language premise]] <italic>p.</italic> The challenges of [[NLI]] are quite different from those encountered in [[formal deduction]]: the emphasis is on [[informal reasoning]], [[lexical semantic knowledge]], and [[variability]] of [[linguistic expression.This dissertation]] explores a range of [[approaches to NLI]], beginning with [[method]]s which are robust but [[approximate]], and proceeding to progressively more precise [[approach]]es. </s>
Line 31: Line 31:
 
Finally, [[we]] address the [[problem of alignment]] for [[NLI]], by developing a [[model]] of [[phrase-based alignment]] inspired by [[analogous work]] in [[machine translation]], including an [[alignment scoring function]], [[inference algorithm]]s for finding [[good alignment]]s, and [[training algorithm]]s for choosing [[feature weight]]s. </s>
 
Finally, [[we]] address the [[problem of alignment]] for [[NLI]], by developing a [[model]] of [[phrase-based alignment]] inspired by [[analogous work]] in [[machine translation]], including an [[alignment scoring function]], [[inference algorithm]]s for finding [[good alignment]]s, and [[training algorithm]]s for choosing [[feature weight]]s. </s>
  
==References==
+
== References ==
 
{{#ifanon:|
 
{{#ifanon:|
 
* 1. Maria Moise, Ciprian Gheorghe, Marilena Zingale, Developing Question Answering (QA) Systems Using the Patterns, WSEAS Transactions on Computers, v.9 n.7, p.726-737, July 2010
 
* 1. Maria Moise, Ciprian Gheorghe, Marilena Zingale, Developing Question Answering (QA) Systems Using the Patterns, WSEAS Transactions on Computers, v.9 n.7, p.726-737, July 2010

Revision as of 23:34, 12 September 2019

Subject Headings:

Notes

Cited By

Quotes

Abstract

Inference has been a central topic in artificial intelligence from the start, but while automatic methods for formal deduction have advanced tremendously, comparatively little progress has been made on the [[problem of <italic>natural language inference</italic> (NLI)]], that is, determining whether a natural language hypothesis <italic>h</italic> can justifiably be inferred from a natural language premise <italic>p.</italic> The challenges of NLI are quite different from those encountered in formal deduction: the emphasis is on informal reasoning, lexical semantic knowledge, and variability of linguistic expression.This dissertation explores a range of approaches to NLI, beginning with methods which are robust but approximate, and proceeding to progressively more precise approaches.

We first develop a baseline system based on overlap between bags of words. Despite its extreme simplicity, this model achieves surprisingly good results on a standard NLI evaluation, the PASCAL RTE Challenge. However, its effectiveness is limited by its failure to represent semantic structure.

To remedy this lack, we next introduce the Stanford RTE system, which uses typed dependency trees as a proxy for semantic structure, and seeks a low-cost alignment between [[trees for <italic>p</italic> and <italic>h]], </italic> using a cost model which incorporates both lexical and structural matching costs. This system is typical of a category of approaches to NLI based on approximate graph matching. We argue, however, that such methods work best when the entailment decision is based, not merely on the degree of alignment, but also on [[global features of the aligned 〈<italic>p]], h</italic>〉 pair motivated by semantic theory.

Seeking still greater precision, we devote the largest part of the dissertation to developing an approach to [[NLI based on a model of <italic>natural logic]]. </italic> We greatly extend past work in natural logic, which has focused solely on semantic containment and monotonicity, to incorporate both semantic exclusion and implicativity. Our system decomposes an inference problem into a sequence of atomic edits which transforms <italic>p</italic> into <italic> h</italic>; predicts a lexical entailment relation for each edit using a statistical classifier; propagates these relations upward through a syntax tree according to semantic properties of intermediate nodes; and composes the resulting entailment relations across the edit sequence.

Finally, we address the problem of alignment for NLI, by developing a model of phrase-based alignment inspired by analogous work in machine translation, including an alignment scoring function, inference algorithms for finding good alignments, and training algorithms for choosing feature weights.

References

;

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2009 NaturalLanguageInferenceBill MaccartneyNatural Language Inference2009
AuthorBill Maccartney +
titleNatural Language Inference +
year2009 +