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

@inproceedings{Patwa2021OverviewOC,
  title={Overview of CONSTRAINT 2021 Shared Tasks: Detecting English COVID-19 Fake News and Hindi Hostile Posts},
  author={Parth Patwa and Mohit Bhardwaj and Vineeth Guptha and Gitanjali Kumari and Shivam Sharma and Srinivas Pykl and Amitava Das and Asif Ekbal and Shad Akhtar and Tanmoy Chakraborty},
  booktitle={CONSTRAINT@AAAI},
  year={2021}
}
Fake news, hostility, defamation are some of the biggest problems faced in social media We present the findings of the shared tasks (https://constraint-shared-task-2021 github io/ ) conducted at the CONSTRAINT Workshop at AAAI 2021 The shared tasks are ‘COVID19 Fake News Detection in English’ and ‘Hostile Post Detection in Hindi’ The tasks attracted 166 and 44 team submissions respectively The most successful models were BERT or its variations © 2021, Springer Nature Switzerland AG 
Constraint 2021: Machine Learning Models for COVID-19 Fake News Detection Shared Task
TLDR
This system addresses the challenge of classifying COVID-19 related social media posts as either fake or real by applying classical machine learning algorithms together with several linguistic features, such as n-grams, readability, emotional tone and punctuation. Expand
Exploring Text-transformers in AAAI 2021 Shared Task: COVID-19 Fake News Detection in English
TLDR
This paper proposed an ensemble method of different pre-trained language models such as BERT, Roberta, Ernie, etc with various training strategies including warm-up, learning rate schedule and k-fold cross-validation for COVID-19 Fake News Detection in English. Expand
Fake News Detection System using XLNet model with Topic Distributions: CONSTRAINT@AAAI2021 Shared Task
TLDR
The team introduced an approach to combine topical distributions from Latent Dirichlet Allocation (LDA) with contextualized representations from XLNet, and compared the method with existing baselines to show that XLNet \(+\) Topic Distributions outperforms other approaches by attaining an F1-score of 0.967. Expand
Hostility Detection and Covid-19 Fake News Detection in Social Media
TLDR
This paper addresses the problem of detecting hostile and fake content in the Devanagari (Hindi) script as a multi-class, multilabel problem using NLP techniques, and builds a model that makes use of an abusive language detector coupled with features extracted via Hindi BERT and Hindi FastText models and metadata. Expand
A Poster on Learnings from an Attempt to Build an NLP-based Fake News Classification system for Hindi
TLDR
This exploratory study curates a dataset of around 13,000 data points of true news articles, and, articles on fake news authored by media organisations which flag fake news and uses seven ML classification models on this dataset and presents the preliminary results. Expand
Extracting Latent Information from Datasets in CONSTRAINT 2021 Shared Task
TLDR
The result of Team Grenzlinie’s experiment in CONSTRAINT 2021 shared task chooses RoBERTa as the pre-trained model, and tries to build a graph from news datasets, achieving an accuracy of 98 64% and an F1-score of98 64% on the test dataset. Expand
A Heuristic-driven Ensemble Framework for COVID-19 Fake News Detection
TLDR
This paper describes a Fake News Detection system that automatically identifies whether a tweet related to COVID-19 is “real” or “fake”, as a part of CONSTRAINT COVID19 Fake News detection in English challenge, and uses an ensemble model consisting of pre-trained models to achieve a joint 8th position on the leader board. Expand
Divide and Conquer: An Ensemble Approach for Hostile Post Detection in Hindi
TLDR
A two level architecture which is made up of BERT based classifiers and statistical classifiers to solve the multi-label multi-class classification problem of Hostile Post Detection in Hindi. Expand
Walk in Wild: An Ensemble Approach for Hostility Detection in Hindi Posts
TLDR
This paper develops a simple ensemble based model on pre-trained mBERT and popular classification algorithms like Artificial Neural Network (ANN) and XGBoost for hostility detection in Hindi posts and received third overall rank in the competition. Expand
Identifying COVID-19 Fake News in Social Media
TLDR
This work trains models which can identify health news related to COVID-19 pandemic as real or fake, and achieves second place on the leaderboard with a very narrow margin 0.05% points. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 61 REFERENCES
Exploring Text-transformers in AAAI 2021 Shared Task: COVID-19 Fake News Detection in English
TLDR
This paper proposed an ensemble method of different pre-trained language models such as BERT, Roberta, Ernie, etc with various training strategies including warm-up, learning rate schedule and k-fold cross-validation for COVID-19 Fake News Detection in English. Expand
Hostility Detection Dataset in Hindi
TLDR
A novel hostility detection dataset in Hindi language collected and manually annotate ~8200 online posts that covers four hostility dimensions: fake news, hate speech, offensive, and defamation posts, along with a non-hostile label. Expand
Fake News Detection System using XLNet model with Topic Distributions: CONSTRAINT@AAAI2021 Shared Task
TLDR
The team introduced an approach to combine topical distributions from Latent Dirichlet Allocation (LDA) with contextualized representations from XLNet, and compared the method with existing baselines to show that XLNet \(+\) Topic Distributions outperforms other approaches by attaining an F1-score of 0.967. Expand
Hostility Detection and Covid-19 Fake News Detection in Social Media
TLDR
This paper addresses the problem of detecting hostile and fake content in the Devanagari (Hindi) script as a multi-class, multilabel problem using NLP techniques, and builds a model that makes use of an abusive language detector coupled with features extracted via Hindi BERT and Hindi FastText models and metadata. Expand
Combating Hostility: Covid-19 Fake News and Hostile Post Detection in Social Media
TLDR
This paper illustrates a detail description of the system and its results that developed as a part of the participation at CONSTRAINT shared task in AAAI-2021, a binary classification problem with fake and real class, while task-B is a multi-label multi-class classification task with five hostile classes. Expand
Extracting Latent Information from Datasets in CONSTRAINT 2021 Shared Task
TLDR
The result of Team Grenzlinie’s experiment in CONSTRAINT 2021 shared task chooses RoBERTa as the pre-trained model, and tries to build a graph from news datasets, achieving an accuracy of 98 64% and an F1-score of98 64% on the test dataset. Expand
A Heuristic-driven Ensemble Framework for COVID-19 Fake News Detection
TLDR
This paper describes a Fake News Detection system that automatically identifies whether a tweet related to COVID-19 is “real” or “fake”, as a part of CONSTRAINT COVID19 Fake News detection in English challenge, and uses an ensemble model consisting of pre-trained models to achieve a joint 8th position on the leader board. Expand
TUDublin team at Constraint@AAAI2021 - COVID19 Fake News Detection
TLDR
The main goal of the work was to create a model that would carry out a binary classification of messages from social media as real or fake news in the context of COVID-19. Expand
Fighting an Infodemic: COVID-19 Fake News Dataset
TLDR
A manually annotated dataset of 10,700 social media posts and articles of real and fake news on COVID-19 is curate and released, and four machine learning baselines are benchmarked. Expand
Divide and Conquer: An Ensemble Approach for Hostile Post Detection in Hindi
TLDR
A two level architecture which is made up of BERT based classifiers and statistical classifiers to solve the multi-label multi-class classification problem of Hostile Post Detection in Hindi. Expand
...
1
2
3
4
5
...