Experimenting a "General purpose" textual entailment learner in AVE

Authors:
Fabio Massimo Zanzotto;Alessandro Moschitti
Affiliations:
ART Group, DISP, University of Rome "Tor Vergata", Rome, Italy;ART Group, DISP, University of Rome "Tor Vergata", Rome, Italy
Venue:
CLEF'06 Proceedings of the 7th international conference on Cross-Language Evaluation Forum: evaluation of multilingual and multi-modal information retrieval
Year:
2006

Citing 8
Cited 1

WordNet: a lexical database for English

Communications of the ACM
Making large-scale support vector machine learning practical

Advances in kernel methods
Applied morphological processing of English

Natural Language Engineering
A maximum-entropy-inspired parser

NAACL 2000 Proceedings of the 1st North American chapter of the Association for Computational Linguistics conference
New ranking algorithms for parsing and tagging: kernels over discrete structures, and the voted perceptron

ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
Automatic learning of textual entailments with cross-pair similarities

ACL-44 Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the Association for Computational Linguistics
WordNet::Similarity: measuring the relatedness of concepts

HLT-NAACL--Demonstrations '04 Demonstration Papers at HLT-NAACL 2004
Measuring the semantic similarity of texts

EMSEE '05 Proceedings of the ACL Workshop on Empirical Modeling of Semantic Equivalence and Entailment

Efficient Graph Kernels for Textual Entailment Recognition

Fundamenta Informaticae - RCRA 2009 Experimental Evaluation of Algorithms for Solving Problems with Combinatorial Explosion

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present the use of a "general purpose" textual entailment recognizer in the Answer Validation Exercise (AVE) task. Our system is designed to learn entailment rules from annotated examples. Its main feature is the use of Support Vector Machines (SVMs) with kernel functions based on cross-pair similarity between entailment pairs. We experimented with our system using different training sets: RTE and AVE data sets. The comparative results show that entailment rules can be learned. Although, the high variability of the outcome prevents us to derive definitive conclusions, the results show that our approach is quite promising and improvable in the future.