Dataset Information

Meta-learning for fake news detection surrounding the Syrian war

ABSTRACT: Summary In this article, we pursue the automatic detection of fake news reporting on the Syrian war using machine learning and meta-learning. The proposed approach is based on a suite of features that include a given article's linguistic style; its level of subjectivity, sensationalism, and sectarianism; the strength of its attribution; and its consistency with other news articles from the same “media camp”. To train our models, we use FA-KES, a fake news dataset about the Syrian war. A suite of basic machine learning models is explored, as well as the model-agnostic meta-learning algorithm (MAML) suitable for few-shot learning, using datasets of a modest size. Feature-importance analysis confirms that the collected features specific to the Syrian war are indeed very important predictors for the output label. The meta-learning model achieves the best performance, improving upon the baseline approaches that are trained exclusively on text features in FA-KES. Highlights • A meta-learning model for distinguishing “true” from “fake” news around the Syrian war• Features capture sectarian discourse and consistency vis-à-vis ground truth• The framework can be extended to detect fake news in other armed conflicts The bigger picture As world and regional powers get more embroiled in the Syrian war, serious questions arise surrounding the credibility of news documenting the facts of war in the decade-long conflict. The spread of fake news around documentation of the war, going beyond plain news bias, not only compromises the integrity of the actual reporting, but also can contribute to psychological warfare that drives the exodus and constant mobility of refugees and hampers humanitarian planning for delivering aid to distraught communities. In this article, we propose meta-learning and machine learning approaches for automatic detection of fake news arising from the Syrian war. Our work reveals the importance of certain features specific to armed conflict in the Middle East, such as sectarian language, and consistency with respect to ground truth from a fact-checking repository. It also highlights the efficacy and quality of meta-learning techniques when tackling datasets of a modest size. We pursue the automatic detection of fake news around the Syrian war using machine learning and meta-learning, using features such as a given article's linguistic style, its level of subjectivity and sectarianism, the strength of its attribution, and its consistency with other articles from the same “media camp”. Our analysis confirms that these features are very important predictors for the output label. Our approach can detect true from fake articles from the FA-KES dataset with extremely high accuracy.

SUBMITTER: Abu Salem F

PROVIDER: S-EPMC8600244 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Introduction In addition to the health crisis that erupted during the COVID-19 pandemic, the war between Russia and Ukraine is impacting the mental health and wellbeing of the Romanian population in a negative way. Objectives This study sets out to investigate the impact that social media consumption and an overload of information related to the armed conflict between Russia and Ukraine is having on the distribution of fake news among Romanians. In addition, it explores the way in which several psychological features, including resilience, general health, perceived stress, coping strategies, and fear of war, change as a function of exposure to traumatic events or interaction with victims of war. Methods Participants (N = 633) completed the General Health Questionnaire (GHQ), the CERQ scale with its nine subscales, the Perceived Stress Scale (PSS), and the BRS scale (Brief Resilience Scale), the last of which measures resilience. Information overload, information strain and the likelihood of the person concerned spreading fake news were assessed by adapting items related to these variables. Findings Our results suggest that information strain partially moderates the relationship between information overload and the tendency to spread false information. Also, they indicate that information strain partially moderates the relationship between time spent online and the tendency to spread false information. Furthermore, our findings imply that there are differences of high and moderate significance between those who worked with refugees and those who did not as regards fear of war and coping strategies. We found no practical differences between the two groups as regards general health, level of resilience and perceived stress. Conclusion and recommendations The importance of discovering the reasons why people share false information is discussed, as is the need to adopt strategies to combat this behavior, including infographics and games designed to teach people how to detect fake news. At the same time, aid workers need to be further supported to maintain a high level of psychological wellbeing.

Project description:BackgroundProliferation of misinformation in digital news environments can harm society in a number of ways, but its dangers are most acute when citizens believe that false news is factually accurate. A recent wave of empirical research focuses on factors that explain why people fall for the so-called fake news. In this scoping review, we summarize the results of experimental studies that test different predictors of individuals' belief in misinformation.MethodsThe review is based on a synthetic analysis of 26 scholarly articles. The authors developed and applied a search protocol to two academic databases, Scopus and Web of Science. The sample included experimental studies that test factors influencing users' ability to recognize fake news, their likelihood to trust it or intention to engage with such content. Relying on scoping review methodology, the authors then collated and summarized the available evidence.ResultsThe study identifies three broad groups of factors contributing to individuals' belief in fake news. Firstly, message characteristics-such as belief consistency and presentation cues-can drive people's belief in misinformation. Secondly, susceptibility to fake news can be determined by individual factors including people's cognitive styles, predispositions, and differences in news and information literacy. Finally, accuracy-promoting interventions such as warnings or nudges priming individuals to think about information veracity can impact judgements about fake news credibility. Evidence suggests that inoculation-type interventions can be both scalable and effective. We note that study results could be partly driven by design choices such as selection of stimuli and outcome measurement.ConclusionsWe call for expanding the scope and diversifying designs of empirical investigations of people's susceptibility to false information online. We recommend examining digital platforms beyond Facebook, using more diverse formats of stimulus material and adding a comparative angle to fake news research.

Dataset Information

Meta-learning for fake news detection surrounding the Syrian war

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets