A Deep Paraphrase Identification Model Interacting Semantics with Syntax

Leilei Kong,Yong Han,Zhongyuan Han,Haoliang Qi,Abd E I.-Baset Hassanien

doi:10.1155/2020/9757032

Abstract

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.

Highlights

Paraphrase is usually done by word replacement with synonyms/antonyms, syntactic modification, sentence reduction, combination, reorganization, word shuffling, concept generalization, and specificity to change the appearance of the original text while retaining the semantics of the source sentence [11], which makes the above methods difficult to further improve the performances using only word matching or vector similarity in the word space
We observe that the DPIM-ISS outperforms the Jaccard approach, the Cosine approach, and the METEOR approach on F1 score and accuracy
We present the DPIM-ISS, a novel text deep paraphrase identification model interacting semantics with syntax

Summary

Introduction

Paraphrase is usually done by word replacement with synonyms/antonyms, syntactic modification, sentence reduction, combination, reorganization, word shuffling, concept generalization, and specificity to change the appearance of the original text while retaining the semantics of the source sentence [11], which makes the above methods difficult to further improve the performances using only word matching or vector similarity in the word space. Ese studies assume that similar texts have similar syntactic structures [12, 15]. A variety of deep models have been introduced into the Complexity research field of paraphrase identification [19,20,21,22,23,24]. Ese models utilized the distributed representation of text and focused on identifying the paraphrase through learning the matching structures and the matching degrees. Except for the widely accepted distributed semantic representation in the deep paraphrase identification models, researchers paid attention to the role of syntax in representing the text and computing the semantic similarity, and proposed some deep paraphrase identification models integrating the syntax [16, 25]. Ese studies determined the validity of syntactic features in deep paraphrase identification Except for the widely accepted distributed semantic representation in the deep paraphrase identification models, researchers paid attention to the role of syntax in representing the text and computing the semantic similarity, and proposed some deep paraphrase identification models integrating the syntax [16, 25]. ese studies determined the validity of syntactic features in deep paraphrase identification

Methods

Results

Conclusion