Research Article
Paraphrase Recognition using Predicate Argument Structure Representation
@INPROCEEDINGS{10.4108/eai.7-12-2021.2314719, author={Chitra A and Anupriya Rajkumar}, title={Paraphrase Recognition using Predicate Argument Structure Representation}, proceedings={Proceedings of the First International Conference on Combinatorial and Optimization, ICCAP 2021, December 7-8 2021, Chennai, India}, publisher={EAI}, proceedings_a={ICCAP}, year={2021}, month={12}, keywords={paraphrase recognition predicate argument matching support vector machine}, doi={10.4108/eai.7-12-2021.2314719} }- Chitra A
Anupriya Rajkumar
Year: 2021
Paraphrase Recognition using Predicate Argument Structure Representation
ICCAP
EAI
DOI: 10.4108/eai.7-12-2021.2314719
Abstract
One of the tasks that make Natural Language Processing applications challenging is Paraphrase Recognition which is the establishment of semantic equivalence between two text units. A popular approach adopted in Paraphrase Recognition, is the usage of symbolic meaning representations as intermediate forms. In this work, Predicate Argument Structures (PAS) have been explored for the task of Paraphrase Recognition. The performance of the system was evaluated on the Microsoft Research Paraphrase Corpus and was found to be superior to existing approaches when the PAS based system was enhanced by using a table of equivalent phrases.
Copyright © 2021–2024 EAI
