A Deep Paraphrase Identification Model Interacting Semantics with Syntax

Paraphrase identification is central to many natural language applications. Based on the insight that a successful paraphrase identification model needs to adequately capture the semantics of the language objects as well as their interactions, we present a deep paraphrase identification model intera...

Full description

Saved in:
Bibliographic Details
Main Authors: Leilei Kong, Zhongyuan Han, Yong Han, Haoliang Qi
Format: Article
Language:English
Published: Wiley 2020-01-01
Series:Complexity
Online Access:http://dx.doi.org/10.1155/2020/9757032
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849397863625261056
author Leilei Kong
Zhongyuan Han
Yong Han
Haoliang Qi
author_facet Leilei Kong
Zhongyuan Han
Yong Han
Haoliang Qi
author_sort Leilei Kong
collection DOAJ
description Paraphrase identification is central to many natural language applications. Based on the insight that a successful paraphrase identification model needs to adequately capture the semantics of the language objects as well as their interactions, we present a deep paraphrase identification model interacting semantics with syntax (DPIM-ISS) for paraphrase identification. DPIM-ISS introduces the linguistic features manifested in syntactic features to produce more explicit structures and encodes the semantic representation of sentence on different syntactic structures by means of interacting semantics with syntax. Then, DPIM-ISS learns the paraphrase pattern from this representation interacting the semantics with syntax by exploiting a convolutional neural network with convolution-pooling structure. Experiments are conducted on the corpus of Microsoft Research Paraphrase (MSRP), PAN 2010 corpus, and PAN 2012 corpus for paraphrase plagiarism detection. The experimental results demonstrate that DPIM-ISS outperforms the classical word-matching approaches, the syntax-similarity approaches, the convolution neural network-based models, and some deep paraphrase identification models.
format Article
id doaj-art-a7bc453cfb7f46a4b33c505249a69d17
institution Kabale University
issn 1076-2787
1099-0526
language English
publishDate 2020-01-01
publisher Wiley
record_format Article
series Complexity
spelling doaj-art-a7bc453cfb7f46a4b33c505249a69d172025-08-20T03:38:49ZengWileyComplexity1076-27871099-05262020-01-01202010.1155/2020/97570329757032A Deep Paraphrase Identification Model Interacting Semantics with SyntaxLeilei Kong0Zhongyuan Han1Yong Han2Haoliang Qi3School of Electronic Information Engineering, Foshan University, Foshan 528225, ChinaSchool of Electronic Information Engineering, Foshan University, Foshan 528225, ChinaSchool of Electronic Information Engineering, Foshan University, Foshan 528225, ChinaSchool of Electronic Information Engineering, Foshan University, Foshan 528225, ChinaParaphrase identification is central to many natural language applications. Based on the insight that a successful paraphrase identification model needs to adequately capture the semantics of the language objects as well as their interactions, we present a deep paraphrase identification model interacting semantics with syntax (DPIM-ISS) for paraphrase identification. DPIM-ISS introduces the linguistic features manifested in syntactic features to produce more explicit structures and encodes the semantic representation of sentence on different syntactic structures by means of interacting semantics with syntax. Then, DPIM-ISS learns the paraphrase pattern from this representation interacting the semantics with syntax by exploiting a convolutional neural network with convolution-pooling structure. Experiments are conducted on the corpus of Microsoft Research Paraphrase (MSRP), PAN 2010 corpus, and PAN 2012 corpus for paraphrase plagiarism detection. The experimental results demonstrate that DPIM-ISS outperforms the classical word-matching approaches, the syntax-similarity approaches, the convolution neural network-based models, and some deep paraphrase identification models.http://dx.doi.org/10.1155/2020/9757032
spellingShingle Leilei Kong
Zhongyuan Han
Yong Han
Haoliang Qi
A Deep Paraphrase Identification Model Interacting Semantics with Syntax
Complexity
title A Deep Paraphrase Identification Model Interacting Semantics with Syntax
title_full A Deep Paraphrase Identification Model Interacting Semantics with Syntax
title_fullStr A Deep Paraphrase Identification Model Interacting Semantics with Syntax
title_full_unstemmed A Deep Paraphrase Identification Model Interacting Semantics with Syntax
title_short A Deep Paraphrase Identification Model Interacting Semantics with Syntax
title_sort deep paraphrase identification model interacting semantics with syntax
url http://dx.doi.org/10.1155/2020/9757032
work_keys_str_mv AT leileikong adeepparaphraseidentificationmodelinteractingsemanticswithsyntax
AT zhongyuanhan adeepparaphraseidentificationmodelinteractingsemanticswithsyntax
AT yonghan adeepparaphraseidentificationmodelinteractingsemanticswithsyntax
AT haoliangqi adeepparaphraseidentificationmodelinteractingsemanticswithsyntax
AT leileikong deepparaphraseidentificationmodelinteractingsemanticswithsyntax
AT zhongyuanhan deepparaphraseidentificationmodelinteractingsemanticswithsyntax
AT yonghan deepparaphraseidentificationmodelinteractingsemanticswithsyntax
AT haoliangqi deepparaphraseidentificationmodelinteractingsemanticswithsyntax