Linguistic Indicators in the Identification of Fake News

Marquardt, Dorota

doi:10.17951/ms.2019.3.95-114

Cited by 10 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…También se utilizan mecanismos automáticos como el text mining, o sea, la minería de texto, que es usada para detectar el uso o ausencia de características léxicas, gramaticales y sintácticas, así como aspectos semánticos referenciales o retóricos (Mahyoob et al, 2020;Marquardt, 2019;O'Brien et al, 2018).…”

Section: Plataformas De Fact-checkingunclassified

Estrategias para luchar contra la desinformación en la comunicación corporativa

Gomes Gonçalves,

Vázquez-González,

González-Sánchez

2023

Redmarka Rev Mark Apl

View full text Add to dashboard Cite

La reputación de las organizaciones puede verse afectada por la difusión de fake news y otros contenidos desinformativos. Así, para salvaguardar la imagen corporativa, los responsables de Comunicación deben estar atentos a las diferentes estrategias para luchar contra la desinformación, cómo el fact-checking, la regulación y la alfabetización mediática. En este artículo pretendemos hacer una contextualización teórica de estas estrategias, presentando el estado actual de la cuestión y revisando la literatura emergente de una temática que aún está poco explorada. Con respecto al fact-checking, aunque sigue siendo necesaria intervención humana para evitar descontextualizaciones y exclusión de discursos híbridos, la evolución tecnológica permite el análisis y moderación de contenidos, gracias a la inteligencia artificial y a sofisticados mecanismos automáticos de minería de textos. En materia de regulación, es siempre muy controvertido aprobar leyes que restringen la libertad de expresión e información. Por otro lado, urge una alfabetización mediática y una educomunicación más envolvente, que englobe a toda la comunidad, desde los más jóvenes hasta los mayores.

show abstract

Section: Plataformas De Fact-checkingunclassified

Estrategias para luchar contra la desinformación en la comunicación corporativa

Gomes Gonçalves,

Vázquez-González,

González-Sánchez

2023

Redmarka Rev Mark Apl

View full text Add to dashboard Cite

show abstract

“…Despite the list of problems enumerated above with NLPs, for those engaged in combatting fake news, the ability to employ computer-aided social network analysis offers great potential for mapping (and ideally countering) the spread on misinformation online, by quickly identifying the key vectors of fake news and tracking (in near-real-time) the flow of fake news across the Web (Hardaker and McGlashan, 2016;Chetty and Alathur, 2018). The use of Computational and Computer-Aided Corpus Linguistics (the identification of key textual features through comparison of a target text with a corpus or corpora of reference texts) offers a real possibility to create automated tools for the identification of fake news through its linguistic content and its removal from social media without human intervention and for automated generation of effective counter-texts (Pérez-Rosas et al, 2018;Marquardt, 2019;Pathak and Srihari, 2019).…”

Section: Computational Linguistics-contextualization and Descriptive mentioning

confidence: 99%

Interdisciplinary Lessons Learned While Researching Fake News

et al. 2020

View full text Add to dashboard Cite

The misleading and propagandistic tendencies in American news reporting have been a part of public discussion from its earliest days as a republic (Innis, 2007; Sheppard, 2007). “Fake news” is hardly new (McKernon, 1925), and the term has been applied to a variety of distinct phenomenon ranging from satire to news, which one may find disagreeable (Jankowski, 2018; Tandoc et al., 2018). However, this problem has become increasingly acute in recent years with the Macquarie Dictionary declaring “fake news” the word of the year in 2016 (Lavoipierre, 2017). The international recognition of fake news as a problem (Pomerantsev and Weiss, 2014; Applebaum and Lucas, 2016) has led to a number of initiatives to mitigate perceived causes, with varying levels of success (Flanagin and Metzger, 2014; Horne and Adali, 2017; Sample et al., 2018). The inability to create a holistic solution continues to stymie researchers and vested parties. A significant contributor to the problem is the interdisciplinary nature of digital deception. While technology enables the rapid and wide dissemination of digitally deceptive data, the design and consumption of data rely on a mixture of psychology, sociology, political science, economics, linguistics, marketing, and fine arts. The authors for this effort discuss deception’s history, both old and new, from an interdisciplinary viewpoint and then proceed to discuss how various disciplines contribute to aiding in the detection and countering of fake news narratives. A discussion of various fake news types (printed, staged events, altered photographs, and deep fakes) ensues with the various technologies being used to identify these; the shortcomings of those technologies and finally the insights offered by the other disciplines can be incorporated to improve outcomes. A three-point evaluation model that focuses on contextual data evaluation, pattern spread, and archival analysis of both the author and publication archives is introduced. While the model put forth cannot determine fact from fiction, the ability to measure distance from fact across various domains provides a starting point for evaluating the veracity of a new story.

show abstract

“…Some researchers have considered linguistic characteristics to detect false news. Marquardt, D. [5] stated that a significant difference between the true and fake corpora is in the noun-to-verb ratio. In real news, the ratio is higher at an average of 4.27 compared to the fake news corpus at 2.73.…”

Section: Introductionmentioning

confidence: 99%

Fake News Classification Based on Content Level Features

et al. 2022

View full text Add to dashboard Cite

Due to the openness and easy accessibility of online social media (OSM), anyone can easily contribute a simple paragraph of text to express their opinion on an article that they have seen. Without access control mechanisms, it has been reported that there are many suspicious messages and accounts spreading across multiple platforms. Accordingly, identifying and labeling fake news is a demanding problem due to the massive amount of heterogeneous content. In essence, the functions of machine learning (ML) and natural language processing (NLP) are to enhance, speed up, and automate the analytical process. Therefore, this unstructured text can be transformed into meaningful data and insights. In this paper, the combination of ML and NLP are implemented to classify fake news based on an open, large and labeled corpus on Twitter. In this case, we compare several state-of-the-art ML and neural network models based on content-only features. To enhance classification performance, before the training process, the term frequency-inverse document frequency (TF-IDF) features were applied in ML training, while word embedding was utilized in neural network training. By implementing ML and NLP methods, all the traditional models have greater than 85% accuracy. All the neural network models have greater than 90% accuracy. From the experiments, we found that the neural network models outperform the traditional ML models by, on average, approximately 6% precision, with all neural network models reaching up to 90% accuracy.

show abstract

Linguistic Indicators in the Identification of Fake News

Cited by 10 publications

References 5 publications

Estrategias para luchar contra la desinformación en la comunicación corporativa

Estrategias para luchar contra la desinformación en la comunicación corporativa

Interdisciplinary Lessons Learned While Researching Fake News

Fake News Classification Based on Content Level Features

Contact Info

Product

Resources

About