Stance detection with BERT embeddings for credibility analysis of information on social media

Karande, Hema; Walambe, Rahee; Benjamin, Victor; Kotecha, Ketan; Raghu, T. S.

doi:10.7717/peerj-cs.467

Cited by 25 publications

(10 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It could be because the CNN algorithm might extract local and global features very well from the vectors using the convolutional, the pooling, and the fully connected (dense) layers, which can maintain semantic context meaning on text data. This finding supports studies on sentiment analysis of a commodity review and stance detection for credibility analysis of information on social media conducted by [24], [25]. These studies showed that BERT embeddings and CNN obtained better results than single CNN that ignores relation contextual semantics on text.…”

Section: Discussionsupporting

confidence: 87%

On the Role of Text Preprocessing in BERT Embedding-based DNNs for Classifying Informal Texts

Kurniasih¹,

Manik²

2022

IJACSA

View full text Add to dashboard Cite

Due to highly unstructured and noisy data, analyzing society reports in written texts is very challenging. Classifying informal text data is still considered a difficult task in natural language processing since the texts could contain abbreviated words, repeating characters, typos, slang, et cetera. Therefore, text preprocessing is commonly performed to remove the noises and make the texts more structured. However, we argued that most tasks of preprocessing are no longer required if suitable word embeddings approach and deep neural network (DNN) architecture are correctly chosen. This study investigated the effects of text preprocessing in fine-tuning a pre-trained Bidirectional Encoder Representations from Transformers (BERT) model using various DNN architectures such as multilayer perceptron (MLP), long short-term memory (LSTM), bidirectional long-short term memory (Bi-LSTM), convolutional neural network (CNN), and gated recurrent unit (GRU). Various experiments were conducted using numerous learning rates and batch sizes. As a result, text preprocessing had insignificant effects on most models such as LSTM, Bi-LSTM, and CNN. Moreover, the combination of BERT embeddings and CNN produced the best classification performance.

show abstract

Section: Discussionsupporting

confidence: 87%

On the Role of Text Preprocessing in BERT Embedding-based DNNs for Classifying Informal Texts

Kurniasih¹,

Manik²

2022

IJACSA

View full text Add to dashboard Cite

show abstract

“…Stance polarity and intensity can be used as a feature to solve a multitude of task such as fake news detection (Karande et al , 2021), opinion polarization (Sirrianni et al , 2018), identifying outlier opinions (Arvapally et al , 2017) and rumor detection (Zubiaga et al , 2018). While these task can be performed using only stance polarity, a study on Cyber Argumentation shows that using stance polarity with the intensity of the relationship can improve the results of discussion analysis (Sirrianni et al , 2018).…”

Section: Related Workmentioning

confidence: 99%

“…Instead, to improve BERT performance on classification problem, several works have found success in combining BERT architecture with other deep learning method such as convolutional neural network (CNN), gated recurrent unit (GRU), long–short-term memory (LSTM) and Bi-LSTM. These research used hybrid BERT model for various tasks such as fake news detection (Karande et al , 2021), sentiment analysis (Zheng and Yang, 2019), identifying offensive speech (Safaya et al , 2020) and intent determination (He et al , 2019). In this research, they found that CNN method is able to outperform other deep-learning technique in improving the BERT performance.…”

Section: Introductionmentioning

confidence: 99%

CNN-BERT for measuring agreement between argument in online discussion

Harly

Girsang

2022

IJWIS

View full text Add to dashboard Cite

Purpose 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/approach The 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). Findings Experiment 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/value The 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.

show abstract

“…we evaluated our models using Recall (R), Precision (P), and macro F1-measure. These machine and deep learning classifiers have shown competitive performance for several NLP tasks (Devlin et al, 2019;Kim, 2014;Hochreiter & Schmidhuber, 1997;Breiman, 2001;Kohavi, 1995;Bashir et al, 2019;Khan et al, 2021;Butt et al, 2021b;Karande et al, 2021;Ashraf et al, 2021;Ameer et al, 2021;Butt et al, 2021a).…”

Section: Benchmarksmentioning

confidence: 99%

Abusive language detection in youtube comments leveraging replies as conversational context

Ashraf

Zubiaga

Gelbukh

2021

PeerJ Computer Science

View full text Add to dashboard Cite

Nowadays, social media experience an increase in hostility, which leads to many people suffering from online abusive behavior and harassment. We introduce a new publicly available annotated dataset for abusive language detection in short texts. The dataset includes comments from YouTube, along with contextual information: replies, video, video title, and the original description. The comments in the dataset are labeled as abusive or not and are classified by topic: politics, religion, and other. In particular, we discuss our refined annotation guidelines for such classification. We report a number of strong baselines on this dataset for the tasks of abusive language detection and topic classification, using a number of classifiers and text representations. We show that taking into account the conversational context, namely, replies, greatly improves the classification results as compared with using only linguistic features of the comments. We also study how the classification accuracy depends on the topic of the comment.

show abstract

Stance detection with BERT embeddings for credibility analysis of information on social media

Cited by 25 publications

References 48 publications

On the Role of Text Preprocessing in BERT Embedding-based DNNs for Classifying Informal Texts

On the Role of Text Preprocessing in BERT Embedding-based DNNs for Classifying Informal Texts

CNN-BERT for measuring agreement between argument in online discussion

Abusive language detection in youtube comments leveraging replies as conversational context

Contact Info

Product

Resources

About