An Inference Model for Semantic Entailment in Natural Language


Authors:

Rodrigo de Salvo Braz and Roxana Girju and Vasin Punyakanok and Dan Roth and Mark Sammons

Abstract:

Semantic entailment is the problem of determining if the meaning of a given sentence entails that of another. We present a princi- pled approach to semantic entailment that builds on inducing re-repre- sentations of text snippets into a hierarchical knowledge representation along with an optimization-based inferential mechanism that makes use of it to prove semantic entailment. This paper provides details and anal- ysis of the knowledge representation and knowledge resources issues en- countered. We analyze our system's behavior on the PASCAL text col- lection1 and the PARC collection of question-answer pairs2. This is used to motivate and explain some of the design decisions in our hierarchical knowledge representation, that is centered around a predicate-argument type abstract representation of text.

Citation:

R. de Salvo Braz and R. Girju and V. Punyakanok and D. Roth and M. Sammons, An Inference Model for Semantic Entailment in Natural Language. Machine Learning Challenges, Evaluating Predictive Uncertainty, Visual Object Classification and Recognizing Textual Entailment, First PASCAL Machine Learning Challenges Workshop, Revised Selected Papers  (2006) pp. 261--286

Bibitem:

@article{BGPRS06,
  author = {R. de Salvo Braz and R. Girju and V. Punyakanok and D. Roth and M. Sammons},
  title = {An Inference Model for Semantic Entailment in Natural Language},
  pages = {261--286},
  year = {2006},
  publisher = {Springer},
  editor = {J. Quinonero Candela and I. Dagan and B. Magnini and F. d'Alche-Buc},
  journal = {Machine Learning Challenges, Evaluating Predictive Uncertainty, Visual Object Classification and Recognizing Textual Entailment, First PASCAL Machine Learning Challenges Workshop, Revised Selected Papers},
  volume = {3944},
  url = " http://cogcomp.cs.illinois.edu/papers/BGPRS06.pdf",
  funding = {ARDA,ITR-BI,TRECC,CLUSTER,XPRESSMP},
  projects = {KINDLE,TE,KRNLP},
  comment = {Revised version of PASCAL RTE Challenge. Textual Entailment, integrating NLP text analyis, shallow semantic inference},
}