基于文本语义和社交行为信息融合的讽刺检测方法

doi:10.11959/j.issn.2096-109x.2023059

Abstract

Abstract:

Sarcasm is a complex implicit emotion that poses a significant challenge in sentiment analysis, particularly in social network sentiment analysis.Effective sarcasm detection holds immense practical significance in the analysis of network public opinion.The contradictory nature of sarcastic texts, which exhibit implicit semantics opposite to the real emotions of users, often leads to misclassification by traditional sentiment analysis methods.Moreover, sarcasm in daily communication is often conveyed through non-textual cues such as intonation and demeanor.Consequently, sarcasm detection methods solely relying on text semantics fail to incorporate non-textual information, thereby limiting their effectiveness.To leverage the power of text semantics and social behavior information, a sarcasm text detection method based on heterogeneous graph information fusion was proposed.The approach involved the construction of a heterogeneous information network encompassing users, texts, and emotional words.A graph neural network model was then designed to handle the representations of the heterogeneous graph.The model employed a dual-channel attention mechanism to extract social behavior information, captured the deep semantics of text through emotional subgraphs, and ultimately combined text semantics and social behavior information.Extensive experiments conducted on the Twitter dataset demonstrate the superiority of the proposed method over existing approaches for sarcasm text detection and classification.

Key words: sarcasm detection, graph neural network, heterogeneous information fusion, implicit sentiment analysis

CLC Number:

TP183

Zhaoyang FU, Zhikai CHEN, Li PAN. Sarcasm detection method based on fusion of text semantics and social behavior information[J]. Chinese Journal of Network and Information Security, 2023, 9(4): 134-143.

Figures/Tables 12

符号	含义
V	节点集合， $V = {u_{1}, u_{2}, \dots, t_{1}, t_{2}, \dots, k_{1}, k_{2}, \dots}$
A	节点类型集合，A={U,T,K}
u_i	第i个用户（U）类型的节点
t_i	第i个文本（T）类型的节点
k_i	第i个情感关键词（K）类型的节点
R	边属性集合，R={U-U,U-T,T-T,T-K}
$x_{i}$	一条帖子中第i个单词
$S$	由[x₁,x₂,…,x_n]组成的文本的词向量矩阵
$h_{u}^{0}$	用户节点u_i的初始化特征向量
$h_{t}^{0}$	文本节点t_i的初始化特征向量
$h_{u}$	用户节点u_i完成更新的特征向量
$h_{t}$	文本节点t_i完成更新的特征向量
$h_{f}$	融合特征向量

模型	Acc	F1
CNN	61.45%	61.32%
Att-LSTM	63.45%	63.77%
BERT	74.45%	74.62%
RoBERTa	75.15%	75.21%
MIARN	72.45%	72.57%
ADGCN	77.35%	77.85%
ARGAT	76.44%	76.32%
GAT	83.15%	83.51%
CASCADE	73.61%	73.42%
SNIF-GAT	$85 . 33 %$	$85 . 62 %$

关键词数量	Acc	F1值
i=0	80.87%	80.69%
i=30	83.31%	83.66%
$i = 50$	$85 . 33 %$	$85 . 62 %$
i=80	85.45%	85.51%
i=100	85.62%	85.77%
i=150	85.65%	85.76%
i=200	85.68%	85.78%

注意力机制	Acc	F1值
no attention	78.21%	78.46%
edge	83.73%	83.95%
node	83.45%	83.62%
normal	$85 . 33 %$	$85 . 62 %$

SNIF-GAT层数	Acc	F1值
1	82.21%	82.36%
$2$	$85 . 33 %$	$85 . 62 %$
3	83.15%	83.41%
4	79.19%	79.47%

References 27

[1]	VERMA P , SHUKLA N , SHUKLA A P . Techniques of sarcasm detection:a review[C]// 2021 International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE). 2021: 968-972.
[2]	JIA X , DENG Z , MIN F ,et al. Three-way decisions based feature fusion for chinese irony detection[J]. International Journal of Approximate Reasoning, 2019,113: 324-335.
[3]	CAI Y , CAI H , WAN X . Multi-modal sarcasm detection in twitter with hierarchical fusion model[C]// Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 2019: 2506-2515.
[4]	SHRIVASTAVA M , KUMAR S . A pragmatic and intelligent model for sarcasm detection in social media text[J]. Technology in Society, 2021,64:101489.
[5]	RILOFF E , QADIR A , SURVE P ,et al. Sarcasm as contrast between a positive sentiment and negative situation[C]// Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing. 2013: 704-714.
[6]	BOUAZIZI M , OHTSUKI T O . A pattern-based approach for sarcasm detection on twitter[J]. IEEE Access, 2016,4: 5477-5488.
[7]	AMIR S , WALLACE B C , LYU H ,et al. Modelling context with user embeddings for sarcasm detection in social media[C]// Proceedings of The 20th SIGNLL Conference on Computational Natural Language Learning. 2016: 167-177.
[8]	DAS D , CLARK A J . Sarcasm detection on flickr using a cnn[C]// Proceedings of the 2018 International Conference on Computing and Big Data. 2018: 56-61.
[9]	KUMAR A , NARAPAREDDY V T , SRIKANTH V A ,et al. Sarcasm detection using multi-head attention based bidirectional LSTM[J]. IEEE Access, 2020,8: 6388-6397.
[10]	BARUAH A , DAS K , BARBHUIYA F ,et al. Context-aware sarcasm detection using bert[C]// Proceedings of the Second Workshop on Figurative Language Processing. 2020: 83-87.
[11]	GREGORY H , LI S , MOHAMMADI P ,et al. A transformer approach to contextual sarcasm detection in Twitter[C]// Proceedings of the 2nd Workshop on Figurative Language Processing. 2020: 270-275.
[12]	TAY Y , TUAN L A , HUI S C ,et al. Reasoning with sarcasm by reading in-between[J]. arXiv Preprint arXiv:1805.02856, 2018.
[13]	HAZARIKA D , PORIA S , GORANTLA S ,et al. CASCADE:contextual sarcasm detection in online discussion forums[C]// Proceedings of the 27th International Conference on Computational Linguistics. 2018: 1837-1848.
[14]	KIPF T N , WELLING M . Semi-supervised classification with graph convolutional networks[J]. arXiv Preprint arXiv:1609.02907, 2016.
[15]	陈皓, 易平 . 基于图神经网络的代码漏洞检测方法[J]. 网络与信息安全学报, 2021,7(3): 37-45.
	CHEN H , YI P . Code vulnerability detection method based on graph neural network[J]. Chinese Journal of Network and Information Security, 2021,7(3): 37-45.
[16]	LOU C , LIANG B , GUI L ,et al. Affective dependency graph for sarcasm detection[C]// Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2021: 1844-1849.
[17]	LI G , LIN F , CHEN W ,et al. Affection enhanced relational graph attention network for sarcasm detection[J]. Applied Sciences, 2022,12(7): 3639.
[18]	PLEPI J , FLEK L . Perceived and intended sarcasm detection with graph attention networks[C]// Findings of the Association for Computational Linguistics. 2021: 4746-4753.
[19]	SHMUELI B , KU L W , RAY S . Reactive supervision:a new method for collecting sarcasm data[C]// Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). 2020: 2553-2559.
[20]	CAMBRIA E , LI Y , XING F Z ,et al. SenticNet 6:ensemble application of symbolic and sub-symbolic AI for sentiment analysis[C]// Proceedings of the 29th ACM International Conference on Information ＆ Knowledge Management. 2020: 105-114.
[21]	HALLAC I R , MAKINIST S , AY B ,et al. user2vec:social media user representation based on distributed document embeddings[C]// 2019 International Artificial Intelligence and Data Processing Symposium. 2019: 1-5.
[22]	GOEL A , GAUTAM J , KUMAR S . Real time sentiment analysis of tweets using Naive Bayes[C]// 2016 2nd International Conference on Next Generation Computing Technologies (NGCT). 2016: 257-261.
[23]	RAMCHOUN H , GHANOU Y , ETTAOUIL M ,et al. Multilayer perceptron:architecture optimization and training[R]. 2016.
[24]	ZHANG C , LI Q C , SONG D W.2019 . Aspect-based sentiment classification with aspect-specific graph convolutional networks[C]// Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), 2019: 4568-4578.
[25]	OPREA S , MAGDY W . Exploring author context for detecting intended vs perceived sarcasm[C]// Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 2019: 2854-2859.
[26]	KOLCHINSKI Y A , POTTS C . Representing social media users for sarcasm detection[C]// Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. 2018: 1115-1121.
[27]	YANG T , HU L , SHI C ,et al. HGAT:heterogeneous graph attention networks for semi-supervised short text classification[J]. ACM Transactions on Information Systems (TOIS), 2021,39(3): 1-29.

Metrics

Recommended 0

No Suggested Reading articles found!

Sarcasm detection method based on fusion of text semantics and social behavior information

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 12

References 27

Related Articles 4

Metrics

Recommended 0

[1]	Zhao CAI, Tao JING, Shuang REN. Survey on Ethereum phishing detection technology [J]. Chinese Journal of Network and Information Security, 2023, 9(2): 21-32.
[2]	Dibin SHAN, Xuehui DU, Wenjuan WANG, Aodi LIU, Na WANG. Access control relationship prediction method based on GNN dual source learning [J]. Chinese Journal of Network and Information Security, 2022, 8(5): 40-55.
[3]	Hao CHEN, Ping YI. Code vulnerability detection method based on graph neural network [J]. Chinese Journal of Network and Information Security, 2021, 7(3): 37-45.
[4]	Jinyin CHEN, Dunjie ZHANG, Guohan HUANG, Xiang LIN, Liang BAO. Adversarial attack and defense on graph neural networks: a survey [J]. Chinese Journal of Network and Information Security, 2021, 7(3): 1-28.