Measuring distance-based semantic similarity using meronymy and hyponymy relations

Yuanyuan Cai, Shirui Pan, Ximeng Wang, Hongshu Chen, Xiaoyan Cai, Min Zuo

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The assessment of semantic similarity between lexical terms plays a critical part in semantic-oriented applications for natural language processing and cognitive science. The optimization of calculation models is still a challenging issue for improving the performance of similarity measurement. In this paper, we investigate WordNet-based measures including distance-based, information-based, feature-based and hybrid. Among them, the distance-based measures are considered to have the lowest computational complexity due to simple distance calculation. However, most of existing works ignore the meronymy relation between concepts and the non-uniformity of path distances caused by various semantic relations, in which path distances are simply determined by conceptual hyponymy relation. To solve this problem, we propose a novel model to calculate the path distance between concepts, and also propose a similarity measure which nonlinearly transforms the distance to semantic similarity. In the proposed model, we assign different weights in accordance with various relations to edges that link different concepts. On basis of the distance model, we use five structure properties of WordNet for similarity measurement, which consist of multiple meanings, multiple inheritance, link type, depth and local density. Our similarity measure is compared against state-of-the-art WordNet-based measures on M&C dataset, R&G dataset and WS-353 dataset. According to experiment results, the proposed measure in this work outperforms others in terms of both Pearson and Spearman correlation coefficients, which indicates the effectiveness of our distance model. Besides, we construct six additional benchmarks to prove that the proposed measure maintains stable performance.

Original languageEnglish
Number of pages14
JournalNeural Computing and Applications
DOIs
Publication statusAccepted/In press - 2019
Externally publishedYes

Keywords

  • Path distance
  • Semantic relationship
  • Semantic similarity
  • Structure property
  • WordNet

Cite this

Cai, Yuanyuan ; Pan, Shirui ; Wang, Ximeng ; Chen, Hongshu ; Cai, Xiaoyan ; Zuo, Min . / Measuring distance-based semantic similarity using meronymy and hyponymy relations. In: Neural Computing and Applications. 2019.
@article{7683e6720e6b42d9bd21d8568b72b60f,
title = "Measuring distance-based semantic similarity using meronymy and hyponymy relations",
abstract = "The assessment of semantic similarity between lexical terms plays a critical part in semantic-oriented applications for natural language processing and cognitive science. The optimization of calculation models is still a challenging issue for improving the performance of similarity measurement. In this paper, we investigate WordNet-based measures including distance-based, information-based, feature-based and hybrid. Among them, the distance-based measures are considered to have the lowest computational complexity due to simple distance calculation. However, most of existing works ignore the meronymy relation between concepts and the non-uniformity of path distances caused by various semantic relations, in which path distances are simply determined by conceptual hyponymy relation. To solve this problem, we propose a novel model to calculate the path distance between concepts, and also propose a similarity measure which nonlinearly transforms the distance to semantic similarity. In the proposed model, we assign different weights in accordance with various relations to edges that link different concepts. On basis of the distance model, we use five structure properties of WordNet for similarity measurement, which consist of multiple meanings, multiple inheritance, link type, depth and local density. Our similarity measure is compared against state-of-the-art WordNet-based measures on M&C dataset, R&G dataset and WS-353 dataset. According to experiment results, the proposed measure in this work outperforms others in terms of both Pearson and Spearman correlation coefficients, which indicates the effectiveness of our distance model. Besides, we construct six additional benchmarks to prove that the proposed measure maintains stable performance.",
keywords = "Path distance, Semantic relationship, Semantic similarity, Structure property, WordNet",
author = "Yuanyuan Cai and Shirui Pan and Ximeng Wang and Hongshu Chen and Xiaoyan Cai and Min Zuo",
year = "2019",
doi = "10.1007/s00521-018-3766-9",
language = "English",
journal = "Neural Computing and Applications",
issn = "0941-0643",
publisher = "Springer-Verlag London Ltd.",

}

Measuring distance-based semantic similarity using meronymy and hyponymy relations. / Cai, Yuanyuan; Pan, Shirui; Wang, Ximeng; Chen, Hongshu ; Cai, Xiaoyan; Zuo, Min .

In: Neural Computing and Applications, 2019.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Measuring distance-based semantic similarity using meronymy and hyponymy relations

AU - Cai, Yuanyuan

AU - Pan, Shirui

AU - Wang, Ximeng

AU - Chen, Hongshu

AU - Cai, Xiaoyan

AU - Zuo, Min

PY - 2019

Y1 - 2019

N2 - The assessment of semantic similarity between lexical terms plays a critical part in semantic-oriented applications for natural language processing and cognitive science. The optimization of calculation models is still a challenging issue for improving the performance of similarity measurement. In this paper, we investigate WordNet-based measures including distance-based, information-based, feature-based and hybrid. Among them, the distance-based measures are considered to have the lowest computational complexity due to simple distance calculation. However, most of existing works ignore the meronymy relation between concepts and the non-uniformity of path distances caused by various semantic relations, in which path distances are simply determined by conceptual hyponymy relation. To solve this problem, we propose a novel model to calculate the path distance between concepts, and also propose a similarity measure which nonlinearly transforms the distance to semantic similarity. In the proposed model, we assign different weights in accordance with various relations to edges that link different concepts. On basis of the distance model, we use five structure properties of WordNet for similarity measurement, which consist of multiple meanings, multiple inheritance, link type, depth and local density. Our similarity measure is compared against state-of-the-art WordNet-based measures on M&C dataset, R&G dataset and WS-353 dataset. According to experiment results, the proposed measure in this work outperforms others in terms of both Pearson and Spearman correlation coefficients, which indicates the effectiveness of our distance model. Besides, we construct six additional benchmarks to prove that the proposed measure maintains stable performance.

AB - The assessment of semantic similarity between lexical terms plays a critical part in semantic-oriented applications for natural language processing and cognitive science. The optimization of calculation models is still a challenging issue for improving the performance of similarity measurement. In this paper, we investigate WordNet-based measures including distance-based, information-based, feature-based and hybrid. Among them, the distance-based measures are considered to have the lowest computational complexity due to simple distance calculation. However, most of existing works ignore the meronymy relation between concepts and the non-uniformity of path distances caused by various semantic relations, in which path distances are simply determined by conceptual hyponymy relation. To solve this problem, we propose a novel model to calculate the path distance between concepts, and also propose a similarity measure which nonlinearly transforms the distance to semantic similarity. In the proposed model, we assign different weights in accordance with various relations to edges that link different concepts. On basis of the distance model, we use five structure properties of WordNet for similarity measurement, which consist of multiple meanings, multiple inheritance, link type, depth and local density. Our similarity measure is compared against state-of-the-art WordNet-based measures on M&C dataset, R&G dataset and WS-353 dataset. According to experiment results, the proposed measure in this work outperforms others in terms of both Pearson and Spearman correlation coefficients, which indicates the effectiveness of our distance model. Besides, we construct six additional benchmarks to prove that the proposed measure maintains stable performance.

KW - Path distance

KW - Semantic relationship

KW - Semantic similarity

KW - Structure property

KW - WordNet

UR - http://www.scopus.com/inward/record.url?scp=85055524937&partnerID=8YFLogxK

U2 - 10.1007/s00521-018-3766-9

DO - 10.1007/s00521-018-3766-9

M3 - Article

JO - Neural Computing and Applications

JF - Neural Computing and Applications

SN - 0941-0643

ER -