A social media analytics platform visualising the spread of COVID-19 in Italy via exploitation of automatically geotagged tweets DOI Creative Commons
Stelios Andreadis, Gerasimos Antzoulatos,

Thanassis Mavropoulos

и другие.

Online Social Networks and Media, Год журнала: 2021, Номер 23, С. 100134 - 100134

Опубликована: Апрель 30, 2021

Social media play an important role in the daily life of people around globe and users have emerged as active part news distribution well production. The threatening pandemic COVID-19 has been lead subject online discussions posts, resulting to large amounts related social data, which can be utilised reinforce crisis management several ways. Towards this direction, we propose a novel framework collect, analyse, visualise Twitter tailored specifically monitor virus spread severely affected Italy. We present evaluate deep learning localisation technique that geotags posts based on locations mentioned their text, face detection algorithm estimate number appearing posted images, community approach identify communities users. Moreover, further analysis collected predict reliability detect trending topics events. Finally, demonstrate platform comprises interactive map display filter analysed utilising outcome technique, visual analytics dashboard visualises results topic, community, event methodologies.

Язык: Английский

Fake or real news about COVID-19? Pretrained transformer model to detect potential misleading news DOI Open Access
Sree Jagadeesh Malla,

P. J. A. Alphonse

The European Physical Journal Special Topics, Год журнала: 2022, Номер 231(18-20), С. 3347 - 3356

Опубликована: Янв. 13, 2022

Язык: Английский

Процитировано

33
Combating the infodemic: COVID-19 induced fake news recognition in social media networks DOI Creative Commons
Shankar Biradar, Sunil Saumya, Arun Chauhan

и другие.

Complex & Intelligent Systems, Год журнала: 2022, Номер 9(3), С. 2879 - 2891

Опубликована: Фев. 18, 2022

Abstract COVID-19 has caused havoc globally due to its transmission pace among the inhabitants and prolific rise in number of people contracting disease worldwide. As a result, seeking information about epidemic via Internet media increased. The impact hysteria that prevailed makes believe share everything related illness without questioning truthfulness. it amplified misinformation spread on social networks disease. Today, there is an immediate need restrict disseminating false news, even more than ever before. This paper presents early fusion-based method for combining key features extracted from context-based embeddings such as BERT, XLNet, ELMo enhance context semantic collection posts achieve higher accuracy news identification. From observation, we found proposed outperforms models work single embeddings. We also conducted detailed studies using several machine learning deep classify platforms relevant COVID-19. To facilitate our work, have utilized dataset “CONSTRAINT shared task 2021” . Our research shown language ensemble are well adapted this role, with 97% accuracy.

Язык: Английский

Процитировано

31
An intelligent cybersecurity system for detecting fake news in social media websites DOI Open Access
Ala Mughaid, Shadi AlZu’bi,

Ahmed AL Arjan

и другие.

Soft Computing, Год журнала: 2022, Номер 26(12), С. 5577 - 5591

Опубликована: Апрель 21, 2022

Язык: Английский

Процитировано

31
Fighting disinformation with artificial intelligence: fundamentals, advances and challenges DOI Creative Commons
Andrés Montoro-Montarroso, Javier Cantón-Correa, Paolo Rosso

и другие.

El Profesional de la Informacion, Год журнала: 2023, Номер unknown

Опубликована: Июнь 17, 2023

Internet and social media have revolutionised the way news is distributed consumed. However, constant flow of massive amounts content has made it difficult to discern between truth falsehood, especially in online platforms plagued with malicious actors who create spread harmful stories. Debunking disinformation costly, which put artificial intelligence (AI) and, more specifically, machine learning (ML) spotlight as a solution this problem. This work revises recent literature on AI ML techniques combat disinformation, ranging from automatic classification feature extraction, well their role creating realistic synthetic content. We conclude that advances been mainly focused scarcely adopted outside research labs due dependence limited-scope datasets. Therefore, efforts should be redirected towards developing AI-based systems are reliable trustworthy supporting humans early detection instead fully automated solutions.

Язык: Английский

Процитировано

22
COVID-19 fake news detection: A hybrid CNN-BiLSTM-AM model DOI
Huosong Xia, Yuan Wang, Zuopeng Zhang

и другие.

Technological Forecasting and Social Change, Год журнала: 2023, Номер 195, С. 122746 - 122746

Опубликована: Июль 28, 2023

Язык: Английский

Процитировано

22
CovTiNet: Covid text identification network using attention-based positional embedding feature fusion DOI Creative Commons

Md. Rajib Hossain,

Mohammed Moshiul Hoque, Nazmul Siddique

и другие.

Neural Computing and Applications, Год журнала: 2023, Номер 35(18), С. 13503 - 13527

Опубликована: Март 14, 2023

Covid text identification (CTI) is a crucial research concern in natural language processing (NLP). Social and electronic media are simultaneously adding large volume of Covid-affiliated on the World Wide Web due to effortless access Internet, gadgets outbreak. Most these texts uninformative contain misinformation, disinformation malinformation that create an infodemic. Thus, essential for controlling societal distrust panic. Though very little Covid-related (such as disinformation, misinformation fake news) has been reported high-resource languages (e.g. English), CTI low-resource (like Bengali) preliminary stage date. However, automatic Bengali challenging deficit benchmark corpora, complex linguistic constructs, immense verb inflexions scarcity NLP tools. On other hand, manual arduous costly their messy or unstructured forms. This proposes deep learning-based network (CovTiNet) identify Bengali. The CovTiNet incorporates attention-based position embedding feature fusion text-to-feature representation CNN identification. Experimental results show proposed achieved highest accuracy 96.61±.001% developed dataset (BCovC) compared methods baselines (i.e. BERT-M, IndicBERT, ELECTRA-Bengali, DistilBERT-M, BiLSTM, DCNN, CNN, LSTM, VDCNN ACNN).

Язык: Английский

Процитировано

18
Fake news detection based on dual-channel graph convolutional attention network DOI
Mengfan Zhao, Yutao Zhang, Guozheng Rao

и другие.

The Journal of Supercomputing, Год журнала: 2024, Номер 80(9), С. 13250 - 13271

Опубликована: Март 4, 2024

Язык: Английский

Процитировано

7
Survey on Deep Learning for Misinformation Detection: Adapting to Recent Events, Multilingual Challenges, and Future Visions DOI
Ansam Khraisat,

Manisha Manisha,

Lennon Y. C. Chang

и другие.

Social Science Computer Review, Год журнала: 2025, Номер unknown

Опубликована: Янв. 24, 2025

The proliferation of misinformation in the digital age has emerged as a pervasive and pressing challenge, threatening integrity information dissemination across online platforms. In response to this growing concern, survey paper offers comprehensive analysis landscape detection methodologies. Our delves into intricacies model architectures, feature engineering, data sources, providing insights strengths limitations each approach. Despite significant advancements detection, identifies persistent challenges. accentuates need for adaptive models that can effectively tackle rapidly evolving events, such COVID-19 pandemic. Language adaptability remains another substantial frontier, particularly context low-resource languages like Chinese. Furthermore, it draws attention dearth balanced, multilingual datasets, emphasizing their significance robust training assessment. By addressing emerging challenges offering view, our enriches understanding deep learning techniques detection.

Язык: Английский

Процитировано

1
Ternion: An Autonomous Model for Fake News Detection DOI Creative Commons
Noman Islam, Asadullah Shaikh, Asma Qaiser

и другие.

Applied Sciences, Год журнала: 2021, Номер 11(19), С. 9292 - 9292

Опубликована: Окт. 6, 2021

In recent years, the consumption of social media content to keep up with global news and verify its authenticity has become a considerable challenge. Social enables us easily access anywhere, anytime, but it also gives rise spread fake news, thereby delivering false information. This negative impact on society. Therefore, is necessary determine whether or not spreading over real. will allow for confusion among users be avoided, important in ensuring positive development. paper proposes novel solution by detecting through natural language processing techniques. Specifically, this scheme comprising three steps, namely, stance detection, author credibility verification, machine learning-based classification, news. last stage proposed pipeline, several learning techniques are applied, such as decision trees, random forest, logistic regression, support vector (SVM) algorithms. For study, dataset was taken from Kaggle. The experimental results show an accuracy 93.15%, precision 92.65%, recall 95.71%, F1-score 94.15% algorithm. SVM better than second best classifier, i.e., 6.82%.

Язык: Английский

Процитировано

41
A systematic literature review and existing challenges toward fake news detection models DOI Open Access

Minal Nirav Shah,

Amit Ganatra

Social Network Analysis and Mining, Год журнала: 2022, Номер 12(1)

Опубликована: Ноя. 14, 2022

Язык: Английский

Процитировано

29