Sememe knowledge and auxiliary information enhanced approach for sarcasm detection

Zhiyuan Wen; Lin Gui; Qianlong Wang; Mingyue Guo; Xiaoqi Yu; Jiachen Du; Ruifeng Xu

doi:10.1016/j.ipm.2022.102883

Sememe knowledge and auxiliary information enhanced approach for sarcasm detection

Zhiyuan Wen, Lin Gui, Qianlong Wang, Mingyue Guo, Xiaoqi Yu, Jiachen Du^*, Ruifeng Xu^*

^*Corresponding author for this work

Informatics

Research output: Contribution to journal › Article › peer-review

31 Citations (Scopus)

Abstract

Sarcasm expression is a pervasive literary technique in which people intentionally express the opposite of what is implied. Accurate detection of sarcasm in a text can facilitate the understanding of speakers’ true intentions and promote other natural language processing tasks, especially sentiment analysis tasks. Since sarcasm is a kind of implicit sentiment expression and speakers deliberately confuse the audience, it is challenging to detect sarcasm only by text. Existing approaches based on machine learning and deep learning achieved unsatisfactory performance when handling sarcasm text with complex expression or needing specific background knowledge to understand. Especially, due to the characteristics of the Chinese language itself, sarcasm detection in Chinese is more difficult. To alleviate this dilemma on Chinese sarcasm detection, we propose a sememe and auxiliary enhanced attention neural model, SAAG. At the word level, we introduce sememe knowledge to enhance the representation learning of Chinese words. Sememe is the minimum unit of meaning, which is a fine-grained portrayal of a word. At the sentence level, we leverage some auxiliary information, such as the news title, to learning the representation of the context and background of sarcasm expression. Then, we construct the representation of text expression progressively and dynamically. The evaluation on a sarcasm dateset, consisting of comments on news text, reveals that our proposed approach is effective and outperforms the state-of-the-art models.

Original language	English
Article number	102883
Journal	INFORMATION PROCESSING AND MANAGEMENT
Volume	59
Issue number	3
Early online date	19 Feb 2022
DOIs	https://doi.org/10.1016/j.ipm.2022.102883
Publication status	Published - May 2022

Keywords

Auxiliary information
Sarcasm detection
Sememe knowledge

Access to Document

10.1016/j.ipm.2022.102883

Cite this

@article{89fa5a00ce434d468935a0f7c770f1d9,

title = "Sememe knowledge and auxiliary information enhanced approach for sarcasm detection",

abstract = "Sarcasm expression is a pervasive literary technique in which people intentionally express the opposite of what is implied. Accurate detection of sarcasm in a text can facilitate the understanding of speakers{\textquoteright} true intentions and promote other natural language processing tasks, especially sentiment analysis tasks. Since sarcasm is a kind of implicit sentiment expression and speakers deliberately confuse the audience, it is challenging to detect sarcasm only by text. Existing approaches based on machine learning and deep learning achieved unsatisfactory performance when handling sarcasm text with complex expression or needing specific background knowledge to understand. Especially, due to the characteristics of the Chinese language itself, sarcasm detection in Chinese is more difficult. To alleviate this dilemma on Chinese sarcasm detection, we propose a sememe and auxiliary enhanced attention neural model, SAAG. At the word level, we introduce sememe knowledge to enhance the representation learning of Chinese words. Sememe is the minimum unit of meaning, which is a fine-grained portrayal of a word. At the sentence level, we leverage some auxiliary information, such as the news title, to learning the representation of the context and background of sarcasm expression. Then, we construct the representation of text expression progressively and dynamically. The evaluation on a sarcasm dateset, consisting of comments on news text, reveals that our proposed approach is effective and outperforms the state-of-the-art models.",

keywords = "Auxiliary information, Sarcasm detection, Sememe knowledge",

author = "Zhiyuan Wen and Lin Gui and Qianlong Wang and Mingyue Guo and Xiaoqi Yu and Jiachen Du and Ruifeng Xu",

note = "Funding Information: This work was partially supported by the National Natural Science Foundation of China (61876053, 62006062,62176076), the Shenzhen Foundational Research Funding, China (JCYJ20180507183527919, JCYJ20200109113441941), Joint Lab of Lab of HITSZ and China Merchants Securities. Funding Information: This work was partially supported by the National Natural Science Foundation of China ( 61876053 , 62006062 , 62176076 ), the Shenzhen Foundational Research Funding, China ( JCYJ20180507183527919 , JCYJ20200109113441941 ), Joint Lab of Lab of HITSZ and China Merchants Securities . Publisher Copyright: {\textcopyright} 2022 Elsevier Ltd",

year = "2022",

month = may,

doi = "10.1016/j.ipm.2022.102883",

language = "English",

volume = "59",

journal = "INFORMATION PROCESSING AND MANAGEMENT",

issn = "0306-4573",

publisher = "Elsevier Limited",

number = "3",

}

TY - JOUR

T1 - Sememe knowledge and auxiliary information enhanced approach for sarcasm detection

AU - Wen, Zhiyuan

AU - Gui, Lin

AU - Wang, Qianlong

AU - Guo, Mingyue

AU - Yu, Xiaoqi

AU - Du, Jiachen

AU - Xu, Ruifeng

N1 - Funding Information: This work was partially supported by the National Natural Science Foundation of China (61876053, 62006062,62176076), the Shenzhen Foundational Research Funding, China (JCYJ20180507183527919, JCYJ20200109113441941), Joint Lab of Lab of HITSZ and China Merchants Securities. Funding Information: This work was partially supported by the National Natural Science Foundation of China ( 61876053 , 62006062 , 62176076 ), the Shenzhen Foundational Research Funding, China ( JCYJ20180507183527919 , JCYJ20200109113441941 ), Joint Lab of Lab of HITSZ and China Merchants Securities . Publisher Copyright: © 2022 Elsevier Ltd

PY - 2022/5

Y1 - 2022/5

N2 - Sarcasm expression is a pervasive literary technique in which people intentionally express the opposite of what is implied. Accurate detection of sarcasm in a text can facilitate the understanding of speakers’ true intentions and promote other natural language processing tasks, especially sentiment analysis tasks. Since sarcasm is a kind of implicit sentiment expression and speakers deliberately confuse the audience, it is challenging to detect sarcasm only by text. Existing approaches based on machine learning and deep learning achieved unsatisfactory performance when handling sarcasm text with complex expression or needing specific background knowledge to understand. Especially, due to the characteristics of the Chinese language itself, sarcasm detection in Chinese is more difficult. To alleviate this dilemma on Chinese sarcasm detection, we propose a sememe and auxiliary enhanced attention neural model, SAAG. At the word level, we introduce sememe knowledge to enhance the representation learning of Chinese words. Sememe is the minimum unit of meaning, which is a fine-grained portrayal of a word. At the sentence level, we leverage some auxiliary information, such as the news title, to learning the representation of the context and background of sarcasm expression. Then, we construct the representation of text expression progressively and dynamically. The evaluation on a sarcasm dateset, consisting of comments on news text, reveals that our proposed approach is effective and outperforms the state-of-the-art models.

AB - Sarcasm expression is a pervasive literary technique in which people intentionally express the opposite of what is implied. Accurate detection of sarcasm in a text can facilitate the understanding of speakers’ true intentions and promote other natural language processing tasks, especially sentiment analysis tasks. Since sarcasm is a kind of implicit sentiment expression and speakers deliberately confuse the audience, it is challenging to detect sarcasm only by text. Existing approaches based on machine learning and deep learning achieved unsatisfactory performance when handling sarcasm text with complex expression or needing specific background knowledge to understand. Especially, due to the characteristics of the Chinese language itself, sarcasm detection in Chinese is more difficult. To alleviate this dilemma on Chinese sarcasm detection, we propose a sememe and auxiliary enhanced attention neural model, SAAG. At the word level, we introduce sememe knowledge to enhance the representation learning of Chinese words. Sememe is the minimum unit of meaning, which is a fine-grained portrayal of a word. At the sentence level, we leverage some auxiliary information, such as the news title, to learning the representation of the context and background of sarcasm expression. Then, we construct the representation of text expression progressively and dynamically. The evaluation on a sarcasm dateset, consisting of comments on news text, reveals that our proposed approach is effective and outperforms the state-of-the-art models.

KW - Auxiliary information

KW - Sarcasm detection

KW - Sememe knowledge

UR - http://www.scopus.com/inward/record.url?scp=85124798145&partnerID=8YFLogxK

U2 - 10.1016/j.ipm.2022.102883

DO - 10.1016/j.ipm.2022.102883

M3 - Article

AN - SCOPUS:85124798145

SN - 0306-4573

VL - 59

JO - INFORMATION PROCESSING AND MANAGEMENT

JF - INFORMATION PROCESSING AND MANAGEMENT

IS - 3

M1 - 102883

ER -

Sememe knowledge and auxiliary information enhanced approach for sarcasm detection

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this