Overview of CONSTRAINT 2021 Shared Tasks: Detecting English COVID-19 Fake News and Hindi Hostile Posts

Patwa, Parth; Bhardwaj, Mohit; Guptha, Vineeth; Kumari, Gitanjali; Sharma, Shivam; Pykl, Srinivas; Das, Amitava; Ekbal, Asif; Akhtar, Md. Shad; Chakraborty, Tanmoy

doi:10.1007/978-3-030-73696-5_5

Cited by 78 publications

(36 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The BERT is seen as the leading method in NLP tasks related to the classification of texts and thus the classification of information and the detection of fake news in it. In the article [25], the results of the competition regarding the detection of fake news related to COVID-19 are presented. Despite the fact that participants used many different methods, the most successful models were BERT and its variations [25].…”

Section: Bert Overviewmentioning

confidence: 99%

“…In the article [25], the results of the competition regarding the detection of fake news related to COVID-19 are presented. Despite the fact that participants used many different methods, the most successful models were BERT and its variations [25]. The advantages of the BERT over classic models, also called shallow learning (e.g., SVM, RF, LR, NB), should be sought in deep learning, which allows to learn feature representations directly from the input without too many manual interventions [21].…”

Section: Bert Overviewmentioning

confidence: 99%

See 1 more Smart Citation

Implementation of the BERT-derived architectures to tackle disinformation challenges

Kula

Kozik

Choraś

2021

Neural Comput & Applic

View full text Add to dashboard Cite

Recent progress in the area of modern technologies confirms that information is not only a commodity but can also become a tool for competition and rivalry among governments and corporations, or can be applied by ill-willed people to use it in their hate speech practices. The impact of information is overpowering and can lead to many socially undesirable phenomena, such as panic or political instability. To eliminate the threats of fake news publishing, modern computer security systems need flexible and intelligent tools. The design of models meeting the above-mentioned criteria is enabled by artificial intelligence and, above all, by the state-of-the-art neural network architectures, applied in NLP tasks. The BERT neural network belongs to this type of architectures. This paper presents Transformer-based hybrid architectures applied to create models for detecting fake news.

show abstract

Section: Bert Overviewmentioning

confidence: 99%

Section: Bert Overviewmentioning

confidence: 99%

Implementation of the BERT-derived architectures to tackle disinformation challenges

Kula

Kozik

Choraś

2021

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…Before building a new dataset for training, we aimed at understanding the dataset at hand for the shared task [15]. We looked at different attributes of the data to get a better understanding of the distribution.…”

Section: Dataset Analysismentioning

confidence: 99%

Tackling the Infodemic: Analysis Using Transformer Based Models

Zutshi

Raj

2021

Combating Online Hostile Posts in Regional Languages During Emergency Situation

View full text Add to dashboard Cite

This paper presents how we tackled the COVID 19 Fake News Detection in English subtask in the SHARED TASK@ CON-STRAINT 2021 using RoBERTa. We perform extensive analysis to understand the pattern of the data distribution. To achieve an F1 score of 0.96, we incorporate external sources of misinformation and fine tune multiple state of the art pretrained deep learning models. In the end, we visualise the true and false positives predicted by our model as improvement in future work.

show abstract

“…For the Fake News datasets, we start by collecting the COVID-19 related fake news. The first one is Constraint@AAAI2021 -COVID19 Fake News Detection in English (Patwa et al, 2021). The data are collected from various social media platforms.…”

Section: Fake Newsmentioning

confidence: 99%

A Domain-Independent Holistic Approach to Deception Detection

Shahriar

Mukherjee

Gnawali

2021

Proceedings of the Conference Recent Advances in Natural Language Processing - Deep Learning for Natural Language Processing Me

View full text Add to dashboard Cite

The deception in the text can be of different forms in different domains, including fake news, rumor tweets, and spam emails. Irrespective of the domain, the main intent of the deceptive text is to deceit the reader. Although domain-specific deception detection exists, domain-independent deception detection can provide a holistic picture, which can be crucial to understand how deception occurs in the text. In this paper, we detect deception in a domain-independent setting using deep learning architectures. Our method outperforms the State-of-the-Art (SOTA) performance of most benchmark datasets with an overall accuracy of 93.42% and F1-Score of 93.22%. The domain-independent training allows us to capture subtler nuances of deceptive writing style. Furthermore, we analyze how much in-domain data may be helpful to accurately detect deception, especially for the cases where data may not be readily available to train. Our results and analysis indicate that there may be a universal pattern of deception lying in-between the text independent of the domain, which can create a novel area of research and open up new avenues in the field of deception detection.

show abstract

Overview of CONSTRAINT 2021 Shared Tasks: Detecting English COVID-19 Fake News and Hindi Hostile Posts

Cited by 78 publications

References 36 publications

Implementation of the BERT-derived architectures to tackle disinformation challenges

Implementation of the BERT-derived architectures to tackle disinformation challenges

Tackling the Infodemic: Analysis Using Transformer Based Models

A Domain-Independent Holistic Approach to Deception Detection

Contact Info

Product

Resources

About