Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

CNN-BERT for measuring agreement between argument in online discussion

CNN-BERT for measuring agreement between argument in online discussion With the rise of online discussion and argument mining, methods that are able to analyze arguments become increasingly important. A recent study proposed the usage of agreement between arguments to represent both stance polarity and intensity, two important aspects in analyzing arguments. However, this study primarily focused on finetuning bidirectional encoder representations from transformer (BERT) model. The purpose of this paper is to propose convolutional neural network (CNN)-BERT architecture to improve the previous method.Design/methodology/approachThe used CNN-BERT architecture in this paper directly uses the generated hidden representation from BERT. This allows for better use of the pretrained BERT model and makes finetuning the pretrained BERT model optional. The authors then compared the CNN-BERT architecture with the method proposed in the previous study (BERT and Siamese-BERT).FindingsExperiment results demonstrate that the proposed CNN-BERT is able to achieve a 71.87% accuracy in measuring agreement between arguments. Compared to the previous study that achieve an accuracy of 68.58%, the CNN-BERT architecture was able to increase the accuracy by 3.29%. The CNN-BERT architecture is also able to achieve a similar result even without further pretraining the BERT model.Originality/valueThe principal originality of this paper is the proposition of using CNN-BERT to better use the pretrained BERT model for measuring agreement between arguments. The proposed method is able to improve performance and also able to achieve a similar result without further training the BERT model. This allows separation of the BERT model from the CNN classifier, which significantly reduces the model size and allows the usage of the same pretrained BERT model for other problems that also did not need to finetune their BERT model. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Web Information Systems Emerald Publishing

CNN-BERT for measuring agreement between argument in online discussion

Download PDF
13 pages

Loading next page...
 
/lp/emerald-publishing/cnn-bert-for-measuring-agreement-between-argument-in-online-discussion-Dw2DhX9DZh
Publisher
Emerald Publishing
Copyright
© Emerald Publishing Limited
ISSN
1744-0084
eISSN
1744-0084
DOI
10.1108/ijwis-12-2021-0141
Publisher site
See Article on Publisher Site

Abstract

With the rise of online discussion and argument mining, methods that are able to analyze arguments become increasingly important. A recent study proposed the usage of agreement between arguments to represent both stance polarity and intensity, two important aspects in analyzing arguments. However, this study primarily focused on finetuning bidirectional encoder representations from transformer (BERT) model. The purpose of this paper is to propose convolutional neural network (CNN)-BERT architecture to improve the previous method.Design/methodology/approachThe used CNN-BERT architecture in this paper directly uses the generated hidden representation from BERT. This allows for better use of the pretrained BERT model and makes finetuning the pretrained BERT model optional. The authors then compared the CNN-BERT architecture with the method proposed in the previous study (BERT and Siamese-BERT).FindingsExperiment results demonstrate that the proposed CNN-BERT is able to achieve a 71.87% accuracy in measuring agreement between arguments. Compared to the previous study that achieve an accuracy of 68.58%, the CNN-BERT architecture was able to increase the accuracy by 3.29%. The CNN-BERT architecture is also able to achieve a similar result even without further pretraining the BERT model.Originality/valueThe principal originality of this paper is the proposition of using CNN-BERT to better use the pretrained BERT model for measuring agreement between arguments. The proposed method is able to improve performance and also able to achieve a similar result without further training the BERT model. This allows separation of the BERT model from the CNN classifier, which significantly reduces the model size and allows the usage of the same pretrained BERT model for other problems that also did not need to finetune their BERT model.

Journal

International Journal of Web Information Systems – Emerald Publishing

Published: Dec 12, 2022

Keywords: Online discussion; Stance polarity and intensity; Natural Language Processing; Applications of Web mining and searching

References

  • Identifying outlier opinions in an online intelligent argumentation system
    Arvapally, R.S.; Liu, X.F.; Nah, F.F.; Jiang, W.
  • Stance detection with bidirectional conditional encoding
    Augenstein, I.; Rocktäschel, T.; Vlachos, A.; Bontcheva, K.
  • BERT: Pre-training of deep bidirectional transformers for language understanding
    Devlin, J.M.; Chang, -W.; Lee, K.; Toutanova, K.
  • The stance triangle
    Du Bois, J.W.
  • But-FIT at SemEval-2019 task 7: determining the rumour stance with pre-trained deep bidirectional transformers
    Fajcik, M.; Smrz, P.; Burget, L.
  • Summarization of Spanish talk shows with Siamese hierarchical attention networks
    González, J.-A.; Hurtado, L.-F.; Segarra, E.; García-Granada, F.; Sanchis, E.
  • A retrospective analysis of the fake news challenge
    Hanselowski, A.; PVS, A.; Schiller, B.; Caspelherr, F.; Chaudhuri, D.; Meyer, C.M.; Gurevych, I.
  • Using convolutional neural network with BERT for intent determination
    He, C.; Chen, S.; Huang, S.; Zhang, J.; Song, X.
  • Stance detection with BERT embeddings for credibility analysis of information on social media
    Karande, H.; Walambe, R.; Benjamin, V.; Kotecha, K.; Raghu, T.
  • Knowledge enhanced masked language model for stance detection
    Kawintiranon, K.; Singh, L.
  • Stance detection: a survey
    Küçük, D.; Can, F.
  • An intelligent computational argumentation system for supporting collaborative software development decision making
    Liu, X.; Khudkhudia, E.; Lei, W.; Sajja, V.
  • SemEval-2016 task 6: detecting stance in tweets
    Mohammad, S.S.; Kiritchenko, P.; Sobhani, X.; Zhu; Cherry, C.
  • Same side stance classification task: facilitating argument stance classification by fine-tuning a BERT model
    Ollinger, S.; Dumani, L.; Sahitaj, P.; Bergmann, R.; Schenkel, R.
  • Online discussion forum as a tool for interactive learning and communication
    Onyema, E.M.; Deborah, E.C.; Alsayed, A.O.; Quadri, N.N.; Sanobe, S.
  • Sentence-BERT: Sentence embeddings using siamese BERT-Network
    Reimers, N.; Gurevych, I.
  • KUISAIL at SemEval-2020 task 12: BERT-CNN for offensive speech identification in social media
    Safaya, A.; Abdullatif, M.; Yuret, D.
  • Agreement prediction of arguments in cyber argumentation for detecting stance polarity and intensity
    Sirrianni, J.; Liu, X.; Adams, D.
  • Quantitative modeling of polarization in online intelligent argumentation and deliberation for capturing collective intelligence
    Sirrianni, J.; Liu, X.; Adams, D.
  • Predicting stance polarity and intensity in cyber argumentation with deep bidirectional transformers
    Sirrianni, J.W.; Liu, X.; Adams, D.
  • Stance detection and analysis in social media
    Sobhani, P.
  • From argumentation mining to stance classification
    Sobhani, P.; Inkpen, D.; Matwin, S.
  • X-Stance: a multilingual multi-target dataset for stance detection
    Vamvas, J.; Sennrich, R.
  • Attention is all you need
    Vaswani, A.N.; Shazeer, N.; Parmar, J.; Uszkoreit, A.N.; Gomez, L.; Kaiser; Polosukhin, I.
  • Google's neural machine translation system: bridging the gap between human and machine translation
    Wu, Y.M.; Schuster, Z.; Chen, Q.V.; Le, M.; Norouzi, W.; Macherey, M.; Krikun, Y.; Cao, Q.; Gao, K.; Macherey, M.; Klingner, J.
  • A new method of improving BERT for text classification
    Zheng, S.; Yang, M.
  • Detection and resolution of rumours in social media: a survey
    Zubiaga, A.; Aker, A.; Bontcheva, K.; Liakata, M.; Procter, R.