Dataset Information

Supervised ensemble learning methods towards automatically filtering Urdu fake news within social media.

ABSTRACT: The popularity of the internet, smartphones, and social networks has contributed to the proliferation of misleading information like fake news and fake reviews on news blogs, online newspapers, and e-commerce applications. Fake news has a worldwide impact and potential to change political scenarios, deceive people into increasing product sales, defaming politicians or celebrities, and misguiding visitors to stop visiting a place or country. Therefore, it is vital to find automatic methods to detect fake news online. In several past studies, the focus was the English language, but the resource-poor languages have been completely ignored because of the scarcity of labeled corpus. In this study, we investigate this issue in the Urdu language. Our contribution is threefold. First, we design an annotated corpus of Urdu news articles for the fake news detection tasks. Second, we explore three individual machine learning models to detect fake news. Third, we use five ensemble learning methods to ensemble the base-predictors' predictions to improve the fake news detection system's overall performance. Our experiment results on two Urdu news corpora show the superiority of ensemble models over individual machine learning models. Three performance metrics balanced accuracy, the area under the curve, and mean absolute error used to find that Ensemble Selection and Vote models outperform the other machine learning and ensemble learning models.

SUBMITTER: Akhter MP

PROVIDER: S-EPMC7959660 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Disinformation (fake news) is a major problem that affects modern populations, especially in an era when information can be spread from one corner of the world to another in just one click. The diffusion of misinformation becomes more problematic when it addresses issues related to health, as it can affect people at both the individual and population levels. Through the ideas proposed by cultural evolution theory, in this study, we seek to understand the dynamics of disseminating messages (cultural traits) with untrue content (maladaptive traits). For our investigation, we used the scenario caused by the Coronavirus Disease 2019 (COVID-19) pandemic as a model. The instability caused by the pandemic provides a good model for the study of adapted and maladaptive traits, as the information can directly affect individual and population fitness. Through data collected on the Twitter platform (259,176 tweets) and using machine learning techniques and web scraping, we built a predictive model to analyze the following questions: (1) Is false information more shared? (2) Is false information more adopted? (3) Do people with social prestige influence the dissemination of maladaptive traits of COVID-19? We observed that fake news features contained in messages with false information were shared and adopted as unblemished messages. We also observed that social prestige was not a determining factor for the diffusion of maladaptive traits. Even with the ability to allow connections between individuals participating in social media, some factors such as attachment to cultural traits and the formation of social bubbles can favor isolation and decrease connectivity between individuals. Consequently, in the scenario of isolation between groups and low connectivity between individuals, there is a reduction in cultural exchange between people, which interferes with the dynamics of the selection of cultural traits. Thus, maladaptive (harmful) traits are favored and maintained in the cultural system. We also argue that the local Brazilian cultural context can be a determining factor for maintaining maladaptive traits. We conclude that in an unstable (pandemic) scenario, the information transmitted on Twitter is not reliable in relation to the increase in fitness, which may occur because of the low cultural exchange promoted by the personalization of the social network and cultural context of the population.Supplementary informationThe online version contains supplementary material available at 10.1007/s42979-021-00836-w.

Project description:IntroductionMisinformation surrounding COVID-19 poses a global public health problem that adversely affects governments' abilities to mitigate the disease and causes accidental deaths and self-harm due to false beliefs about the virus, prevention measures, vaccines and cures. We aim to examine the relationship between exposure to and trust in COVID-19 news (from Television, social media, interpersonal communication) and information sources (healthcare experts, government, clerics) and belief in COVID-19 myths and false information, as well as critical verification practices before posting on social media.MethodsWe use a cross-sectional researcher-administered phone survey of adults living in Lebanon between March 27 and April 23, 2020.ResultsThe sample included 56.1% men and 43.9% women, 37.9% with a university degree, 63.0% older than 30, and 7% with media literacy training. Those who trust COVID-19 news from social media [95%CI:(1.05-1.52)] and interpersonal communication [95%CI:(1.25-1.82)], and those who trust information from clerics [95%CI:(1.25-1.82)] were more likely to believe in COVID-19 myths and false information. University graduates [95%CI:(0.25-0.51)] and those who trust information from government [95%CI:(0.65-0.89] were less likely to believe in myths and false information. Those who believe in COVID-19 myths and false information [95%CI:(0.25-0.70)] were less likely to engage in critical social media posting practices. Only those who underwent media literacy training [95%CI:(1.24-6.55)] were more likely to engage in critical social media posting practices.ConclusionHigher education and trust in information from government contributed to decreasing belief in COVID-19 myths and false information. Trust in news from social media, interpersonal communication and clerics contributed to increasing belief in COVID-19 myths and false information, which in turn contributed to less critical social media posting practices, thereby exacerbated the infodemic. Media literacy training contributed to increasing critical social media posting practices, thereby played a role in mitigating the infodemic.

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

Supervised ensemble learning methods towards automatically filtering Urdu fake news within social media.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets