Cover Image

Chinese sentence similarity calculation based on modifiers

To compute the similarity of Chinese sentences accurately, a revised Chinese sentence similarity approach is proposed though enhancing the importance of the modifiers of stem of sentence. After extracting the modified part of the sentence by Language Technology Platform (LTP), this part of each stru...

Full description

Saved in:
Bibliographic Details
Main Authors: Wang, Fangling, Ye, Shaoqiang, Kang, Diwen, Mohd. Zain, Azlan, Zhou, Kaiqing
Format: Conference or Workshop Item
Published: 2022
Subjects:
QA75 Electronic computers. Computer science
Online Access:http://eprints.utm.my/id/eprint/100495/
http://dx.doi.org/10.1007/978-3-031-06794-5_25
Tags: Add Tag
No Tags, Be the first to tag this record!
id my.utm.100495
record_format eprints
spelling my.utm.1004952023-04-14T02:15:51Z http://eprints.utm.my/id/eprint/100495/ Chinese sentence similarity calculation based on modifiers Wang, Fangling Ye, Shaoqiang Kang, Diwen Mohd. Zain, Azlan Zhou, Kaiqing QA75 Electronic computers. Computer science To compute the similarity of Chinese sentences accurately, a revised Chinese sentence similarity approach is proposed though enhancing the importance of the modifiers of stem of sentence. After extracting the modified part of the sentence by Language Technology Platform (LTP), this part of each structure could be removed the longest common substring, to better capture the similarities of modified parts. The entire method includes three phases, which are to split the sentences into principal and predicate object structures using the syntactic analysis tool, to generate modifiers and sentence stem vectors and calculate the similarity between the vectors using the Word2Vec, and to obtain the similarity between two sentences by weighting each part. Experimental results on 200 sentences of the LCQMC dataset and corresponding analysis reveal that the proposed method can obtain more accurate similarity calculation results by effectively gaining the modified part - which affects the whole sentence meaning effectively-of the sentence structure. 2022 Conference or Workshop Item PeerReviewed Wang, Fangling and Ye, Shaoqiang and Kang, Diwen and Mohd. Zain, Azlan and Zhou, Kaiqing (2022) Chinese sentence similarity calculation based on modifiers. In: 8th International Conference on Artificial Intelligence and Security, ICAIS 2022, 15 - 20 July 2022, Qinghai, China. http://dx.doi.org/10.1007/978-3-031-06794-5_25
institution Universiti Teknologi Malaysia
building UTM Library
collection Institutional Repository
continent Asia
country Malaysia
content_provider Universiti Teknologi Malaysia
content_source UTM Institutional Repository
url_provider http://eprints.utm.my/
topic QA75 Electronic computers. Computer science
spellingShingle QA75 Electronic computers. Computer science
Wang, Fangling
Ye, Shaoqiang
Kang, Diwen
Mohd. Zain, Azlan
Zhou, Kaiqing
Chinese sentence similarity calculation based on modifiers
description To compute the similarity of Chinese sentences accurately, a revised Chinese sentence similarity approach is proposed though enhancing the importance of the modifiers of stem of sentence. After extracting the modified part of the sentence by Language Technology Platform (LTP), this part of each structure could be removed the longest common substring, to better capture the similarities of modified parts. The entire method includes three phases, which are to split the sentences into principal and predicate object structures using the syntactic analysis tool, to generate modifiers and sentence stem vectors and calculate the similarity between the vectors using the Word2Vec, and to obtain the similarity between two sentences by weighting each part. Experimental results on 200 sentences of the LCQMC dataset and corresponding analysis reveal that the proposed method can obtain more accurate similarity calculation results by effectively gaining the modified part - which affects the whole sentence meaning effectively-of the sentence structure.
format Conference or Workshop Item
author Wang, Fangling
Ye, Shaoqiang
Kang, Diwen
Mohd. Zain, Azlan
Zhou, Kaiqing
author_facet Wang, Fangling
Ye, Shaoqiang
Kang, Diwen
Mohd. Zain, Azlan
Zhou, Kaiqing
author_sort Wang, Fangling
title Chinese sentence similarity calculation based on modifiers
title_short Chinese sentence similarity calculation based on modifiers
title_full Chinese sentence similarity calculation based on modifiers
title_fullStr Chinese sentence similarity calculation based on modifiers
title_full_unstemmed Chinese sentence similarity calculation based on modifiers
title_sort chinese sentence similarity calculation based on modifiers
publishDate 2022
url http://eprints.utm.my/id/eprint/100495/
http://dx.doi.org/10.1007/978-3-031-06794-5_25
_version_ 1764222574946942976
score 13.211869

Similar Items