FacTeR-Check: Semi-automated fact-checking through semantic similarity and natural language inference

Martín, Alfonso Jiménez; Huertas‐Tato, Javier; Huertas-García, Álvaro; Villar-Rodríguez, Guillermo; Camacho, David

doi:10.1016/j.knosys.2022.109265

Cited by 26 publications

(20 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In line with this goal, the mechanics in FacTeR-Check (Martín et al, 2021) describe how to extract tweets related to a given claim and filter them according to their position (support or denial) of the claim. The pipeline of this article is inspired by this recent research and comprises the following steps: 1) collection of the selected COVID-19 pieces of false information from fact-checkers; 2) creation of search queries composed of different representative keywords to retrieve tweets from Twitter API (Application Programming Interface; 3) a labelling process through Natural Language Processing (NLP) techniques to determine if every downloaded tweet supports or denies the input claim.…”

Section: Methodsmentioning

confidence: 99%

Virality, only the tip of the iceberg: ways of spread and interaction around COVID-19 misinformation in Twitter

Villar-Rodríguez

Souto‐Rico

Martín

2022

Communication & Society

View full text Add to dashboard Cite

Misinformation has long been a weapon that helps the political, social, and economic interests of different sectors. This became more evident with the transmission of false information in the COVID-19 pandemic, compromising citizens’ health by anti-vaccine recommendations, the denial of the coronavirus and false remedies. Online social networks are the breeding ground for falsehoods and conspiracy theories. Users can share viral misinformation or publish it on their own. This encourages a double analysis of this issue: the need to capture the deluge of false information as opposed to the real one and the study of users’ patterns to interact with that infodemic. As a response to this, our work combines the use of artificial intelligence and journalism through fact-checked false claims to provide an in-depth study of the number of retweets, likes, replies, quotes and repeated texts in posts stating or contradicting misinformation in Twitter. The large sample of tweets was collected and automatically analysed through Natural Language Processing (NLP) techniques, not to give all the attention only to the posts with a big impact but to all the messages contributing to the expansion of false information or its rejection regardless of their virality. This analysis revealed that the diffusion of tweets surrounding coronavirus-related misinformation is not only a domain of viral tweets, but also from posts without interactions, which represent most of the sample, and that there are no big differences between misinformation and its contradiction in general, except for the use of replies.

show abstract

Section: Methodsmentioning

confidence: 99%

Virality, only the tip of the iceberg: ways of spread and interaction around COVID-19 misinformation in Twitter

Villar-Rodríguez

Souto‐Rico

Martín

2022

Communication & Society

View full text Add to dashboard Cite

show abstract

“…In (Martín et al, 2021), the authors developed the pipeline for checking the news on veracity. They compare embeddings of the news under consideration with ones from the database, using cosine distance.…”

Section: Related Workmentioning

confidence: 99%

“…We fine-tune the NLI model on the data of the competition. The approach is based on the one proposed by (Martín et al, 2021).…”

Section: Natural Language Inferencementioning

confidence: 99%

SkoltechNLP at SemEval-2022 Task 8: Multilingual News Article Similarity via Exploration of News Texts to Vector Representations

Kuimov¹,

Dementieva²,

Panchenko³

2022

Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022)

View full text Add to dashboard Cite

This paper describes our contribution to Se-mEval 2022 Task 8 on Multilingual News Article Similarity. The aim was to test completely different approaches and distinguish the best performing. That is why we've considered systems based on Transformer-based encoders, NER-based, and NLI-based methods (and their combination with SVO dependency triplets representation). The results prove that Transformer models produce the best scores. However, there is space for research and approaches that give not yet comparable but more interpretable results. 1 https://huggingface.co/ distilbert-base-multilingual-cased 2 https://huggingface.co/ bert-base-multilingual-cased and https://huggingface.co/ bert-base-multilingual-uncased 3 https://huggingface.co/ xlm-roberta-base and https://huggingface.co/ xlm-roberta-large 4

show abstract

“…Moreover, the use of semantic-aware models has proven to be an excellent approach to counteract informational disorders (i.e. misinformation, disinformation, malinformation, misleading information, or any other kind of information pollution) [ 54 – 57 ] or to build automated fact-checking approaches [ 58 ]. Semantic similarity can be applied to organise data according to text properties, formally an unsupervised thematic analysis [ 59 ].…”

Section: Related Workmentioning

confidence: 99%

Exploring Dimensionality Reduction Techniques in Multilingual Transformers

Huertas-García

Martín²,

Huertas‐Tato³

et al. 2022

Cogn Comput

View full text Add to dashboard Cite

In scientific literature and industry, semantic and context-aware Natural Language Processing-based solutions have been gaining importance in recent years. The possibilities and performance shown by these models when dealing with complex Human Language Understanding tasks are unquestionable, from conversational agents to the fight against disinformation in social networks. In addition, considerable attention is also being paid to developing multilingual models to tackle the language bottleneck. An increase in size has accompanied the growing need to provide more complex models implementing all these features without being conservative in the number of dimensions required. This paper aims to provide a comprehensive account of the impact of a wide variety of dimensional reduction techniques on the performance of different state-of-the-art multilingual siamese transformers, including unsupervised dimensional reduction techniques such as linear and nonlinear feature extraction, feature selection, and manifold techniques. In order to evaluate the effects of these techniques, we considered the multilingual extended version of Semantic Textual Similarity Benchmark (mSTSb) and two different baseline approaches, one using the embeddings from the pre-trained version of five models and another using their fine-tuned STS version. The results evidence that it is possible to achieve an average reduction of $$91.58\% \pm 2.59\%$$ 91.58 % ± 2.59 % in the number of dimensions of embeddings from pre-trained models requiring a fitting time $$96.68\% \pm 0.68\%$$ 96.68 % ± 0.68 % faster than the fine-tuning process. Besides, we achieve $$54.65\% \pm 32.20\%$$ 54.65 % ± 32.20 % dimensionality reduction in embeddings from fine-tuned models. The results of this study will significantly contribute to the understanding of how different tuning approaches affect performance on semantic-aware tasks and how dimensional reduction techniques deal with the high-dimensional embeddings computed for the STS task and their potential for other highly demanding NLP tasks.

show abstract

FacTeR-Check: Semi-automated fact-checking through semantic similarity and natural language inference

Cited by 26 publications

References 24 publications

Virality, only the tip of the iceberg: ways of spread and interaction around COVID-19 misinformation in Twitter

Virality, only the tip of the iceberg: ways of spread and interaction around COVID-19 misinformation in Twitter

SkoltechNLP at SemEval-2022 Task 8: Multilingual News Article Similarity via Exploration of News Texts to Vector Representations

Exploring Dimensionality Reduction Techniques in Multilingual Transformers

Contact Info

Product

Resources

About