The fake news graph analyzer: An open-source software for characterizing spreaders in large diffusion graphs

Machine learning methods are proposed to mitigate the spread of fake Brazilian news about politics so as not to harm society. Supervised algorithms are explored, requiring labeled news to train. However, labeling a high volume of news can be complex, onerous, time-consuming, error-prone, and costly. Hence, large language models (LLMs) have been used to detect fake news once LLMs are unsupervised methods that can play the role of classifiers. Most fake news detection studies explore the OpenAI LLMs (require payment) and lack an empirical evaluation with other LLMs. However, several open-source models obtain comparative and state-of-the-art (SOTA) results. We highlight that these models have yet to be explored in detecting fake Brazilian news about politics, which is crucial as it directly impacts society. In this sense, we propose a new dataset for detecting fake Brazilian news about politics and an empirical evaluation of open-source LLMs and OpenAI LLMs. In our results, the LLM from Google (Gemma) outperformed the other six LLMs, including GPT-4, proving to be the most promising model for detecting fake news about Brazilian politics.

Fake news is the kind of news that is made deliberately to deceive the readers. It is a sort of purposeful publicity which is distributed as veritable information. Fake news is spread out through traditional news media and social media. Fake news has been an issue for quite a while. With the introduction of social media, the spread of fake news is increased, and it got hard to separate between true news and fake news. The spread of fake news involves worry as it manipulates public opinions. The widespread of fake news can adversely affect people and society in general. The problem of fake news has increased significantly in recent years. The scope and impact of social media have shifted dramatically. On the one hand, it is low cost and simple accessibility, as well as the ability to quickly exchange information attract people to read news from it. On the other hand, it empowers widespread fake news, which is only false data to deceive people. A stand-alone bidirectional encoder representations from transformers model is used in this paper for fake news detection. We try to pose the problem as a text classification problem and build a deep learning model for achieving the objective. We used the BERT model to develop fake news or real news classification solution for our solution in this model. We achieved more than 95% accuracy on the test set and a remarkable by the area under the curve (AUC) by a stand-alone BERT model.KeywordsNews detectionBERTDeep learningMachine learningSocial mediaTransformerNLP

Easy access to equipment, software, and platforms to create, distribute, and provide access to fake news stories has exacerbated the problem of fake news, making for a large number of highly biased sources that are reaching the mainstream through social networks. The economics of emotion theory proposes that fake news headlines are created to evoke emotional responses in readers that will cause them to interact with the article in a way that allows the creator to make a profit (through clicking on the link to the full article, by sharing the article, etc.). This mixed methods study investigates the process by which individuals experience discrete emotional reactions to fake news headlines, and how these emotions contribute to the perpetuation of fake news through sharing behaviors. U.S. participants (n=879 across two waves) viewed one of eight false news headlines and reported their emotional reactions, belief in the headline, and potential sharing behaviors. In general, participants were more likely to believe headlines that aligned with their existing beliefs (e.g., liberals were more likely to believe negative news about conservatives), reacted with more negative emotions to headlines that attacked their party, and were more likely to report intentions to suppress (e.g., post a link to a fact check) fake news that attacked their own party. Emotional reactivity of participants was associated with response behavior intentions such that participants who reported high levels of emotions were more likely to take actions that would spread or suppress the fake news, participants who reported low levels of emotions were more likely to ignore or disengage from the spread of false news, and participants who reported high levels of negative emotions and low levels of positive emotions were more likely to suppress the spread of fake news and less likely to contribute to the spread of fake news. Our findings are synthesized into a process model that explains how discrete emotions and beliefs influence sharing behaviors. Implications for mitigating the spread of fake news are discussed in terms of this model.

Elections worldwide have been marked by the spread of fake news. Online disinformation is everywhere, used as a political weapon in the battlefield of manipulation. This study focused on the Portuguese 2019 elections to assess the reach of fake news compared to mainstream media news and to verify whether fake news had specific targets. We reviewed all posts (N=1197) from newspaper Facebook pages and fake news Facebook pages published during the campaign to verify their engagement. BuzzSumo assessed popularity by counting all posts’ shares, reactions and comments. Iramuteq software related the content of all published headlines by analyzing clusters of the most frequent words. Findings show that fake news had no greater reach than real news during the election campaign. However, fake news are more likely to be shared, while real news tend to get more reactions and more comments. In fake news headlines the terms associated with left-wing and government are the most common. The prime minister and the Socialist Party are associated with negative connotations. Results suggest that Portuguese fake news are related to rightwing extremism and publish hate content targeting corruption and leftist policies in general. Unlike other countries, anti-immigration discourse and fearmongering were not prominent contents.

PurposeThe continuous spread of the corona virus disease 2019 (COVID-19) pandemic has generated public health concern with avalanche of information accompanied by series of fake news. Thus, this study examined the effect of Information Literacy Competency (ILC) in curtailing the spread of fake news among Library and Information Science (LIS) undergraduates in Nigeria.Design/methodology/approachSurvey research design of the descriptive type was used to generate data from a group of LIS undergraduates online. The population of the study consisted of 138 LIS undergraduate students who participated in the survey (thus, n = 138). The data retrieved was subjected to descriptive analysis.FindingsThe study revealed that the students had high level of ILC (x ¯ = 3.42), and there was low prevalence level of COVID-19 pandemic fake news (x ¯ = 2.35) among them. The major causes of COVID-19 fake news were too much information in circulation concerning COVID-19 (x ¯ = 3.44) and the resultant inability to discern or spot fake news from verified and authentic news (x ¯ = 3.28). The study also revealed that ILC had a significant effect in curtailing the spread of COVID-19 fake news with a grand mean of 3.28 against the criterion mean of 2.5. It is implied that LIS undergraduates are educationally position to acquire ILC which is crucial to their identification of fake news and helps to curtail its spread.Research limitations/implicationsThe study is limited in its use of online group for data elicitation within a limited period of three weeks. Also, in its adoption of self-evaluation scale to measure ILC instead of standard information literacy test. Also, the high chances of social desirability bias in sections C and E serve as a limitation to the study.Practical implicationsThe study reinforces the need to enhance structures that flags fake news on social media platforms and integrating IL into schools’ curriculum at all levels.Originality/valueThis study seeks to pioneer a new area of focus on the relevance of ILC to different global issues that concern the health and well-being.

Nowadays, with the advancement of technology, the use of news sources has also undergone a great evolution. News sources have constantly evolved from past to present, ranging from magazines to radios, from newspapers to televisions. The fact that it has become so easy to access news has caused society to pay more attention to fake news. Fake news has the ability to influence society through news sources such as social media, which can reach wider audiences with the development of technology. The difficulties of users in accessing accurate and reliable sources in this information flow that shapes their daily lives increases the potential for the spread of fake news, and it becomes increasingly difficult to distinguish between real and fake news. In this study, classification models for fake news detection were designed using machine learning algorithms. The dataset, which includes fake and real news examples, contains 42,000 examples. Each class, including fake and real samples, contains 22,000 sample data. In order to increase data quality, accuracy and usability, preprocessing methods were applied to the data set. The removal of numbers, stop words, and html tags was done in the pre-processing step to remove unnecessary information from the text. Models were created for fake news detection with singular and ensemble classification algorithms. Performance evaluation of the models was performed using 5-fold cross-validation. In the performance comparisons of the models, values such as accuracy, sensitivity, specificity, tp rate and fp rate were calculated. The highest performance results were observed in the random forest classification algorithm with an accuracy rate of 76%.

This study sought to investigate factors causing the spread of fake news on digital media (DM) and to explore the sometimes disastrous consequences of fake news on social media. The study also aimed to construct a framework for fake news disaster management to control the dangers of false news on DM. The study applied PRISMA guidelines and techniques for exploring, devising, and inclusion and exclusion criteria. The search was carried out through 15 of the world’s leading digital databases. As a result, 31 peer-reviewed studies published in impact-factor journals of leading databases were included. Findings showed that several factors influenced the sharing of fake news on digital media (DM) platforms. Six major trending factors were the rise of technologies, social connections, political reasons, the absence of a controlling center, online business and marketing, and quick dissemination of information. The study identified the disadvantages of fake news (FN) on digital media (DM). A framework was constructed for managing fake news disasters to control the spread of fake news on digital media. This paper offers important theoretical contributions through the development of a framework for controlling fake news spread on digital media and by providing a valuable addition to the existing body of knowledge. The study offers practical assistance to top management, decision makers, and policymakers to devise policies to effectively manage problems caused by fake news dissemination. It provides practical strategies to address fake news disasters on digital media for redefining social values. This research also assists digital media managers in utilizing the proposed framework and controlling the harmful impact of fake news on social media.

The online spread of fake news on various platforms has emerged as a significant concern, posing threats to public opinion, political stability, and the dissemination of reliable information. Researchers have turned to advanced technologies, including machine learning (ML) and deep learning (DL) techniques, to detect and classify fake news to address this issue. This research study explores fake news classification using diverse ML and DL approaches. We utilized a well-known “Fake News” dataset sourced from Kaggle, encompassing a labelled news collection. We implemented diverse ML models, including multinomial naïve bayes (MNB), gaussian naïve bayes (GNB), Bernoulli naïve Bayes (BNB), logistic regression (LR), and passive aggressive classifier (PAC). Additionally, we explored DL models, such as long short-term memory (LSTM), convolutional neural networks (CNN), and CNN-LSTM. We compared the performance of these models based on key evaluation metrics, such as accuracy, precision, recall, and the F1 score. Additionally, we conducted cross-validation and hyperparameter tuning to ensure optimal performance. The results provide valuable insights into the strengths and weaknesses of each model in classifying fake news. We observed that DL models, particularly LSTM and CNN-LSTM, showed better performance compared to traditional ML models. These models achieved higher accuracy and demonstrated robustness in classification tasks. These findings emphasize the potential of DL models to tackle the spread of fake news effectively and highlight the importance of utilizing advanced techniques to address this challenging problem.

'Cure or Poison?' Identity Verification and the Spread of Fake News on Social Media

Market Forces: Quantifying the Role of Top Credible Ad Servers in the Fake News Ecosystem

During Coronavirus Disease 2019 (COVID-19) pandemic, the spread of fake news on social media occurred massively. To avoid any negative impacts, people are expected to exhibit investigative behavior towards the news they encounter, including conducting fact checks. Previous research has shown that intellectual humility can influence investigative behavior when exposed to fake news about COVID-19. This study aims to examine intellectual humility’s ability to predict investigative behavior when dealing with news about the COVID-19 vaccine. The study involved 227 students (157 female and 70 male, M = 21, SD = 1.19) as respondents who were selected by convenience sampling. The instruments used in revealing the two variables are the General Intellectual Humility Scale, headlines of news articles about the COVID-19 vaccine, and the scale of investigative behavior tendencies towards news about the COVID-19 vaccine. The results showed that intellectual humility could predict investigative behavior towards fake (B = 0.89; 95% CI [0,62, 1,15], p < 0,001) and fact news headlines (B = 0,87; 95% CI [0,60, 1,15], p < 0,001) about the COVID-19 vaccine. This finding implies that higher intellectual humility in individuals is predicted to increase investigative behavior towards news about the COVID-19 vaccine.

Research background: Social media plays a crucial role today in enhancing or limiting how fake news is spread. Whether devised by man or developed by artificial intelligence, it has the power to rapidly change consumers’ minds, encouraging them to adopt new behaviors, perceive situations differently, or even act in total opposition to what might be expected. The new dynamics of communication highlights the need for an organizational response adapted to new AI technologies and to the dissemination of fake news within social media networks. Purpose of this article: This paper aims to reveal, by means of bibliometric analysis and a systematic literature review, the generative capabilities of artificial intelligence in the creation and spread of fake news in the business environment, acknowledging the role of previous research in predicting accurately the constant developments in contemporary society. Methods: The analysis is based on a PRISMA flowchart to examine how artificial intelligence technologies contribute to the creation of fake news whilst also highlighting potential artificial intelligence regulations and standards for limiting the dissemination of false information. Initially, the database included over 3,400 highly cited articles retrieved from Scopus and Web of Science, published in the last years, from which a total of 203 were selected for inclusion in the analysis. The bibliometric analysis follows research directions related to detection methods and strategies, legislation and policies governing artificial intelligence technologies used in the creation and dissemination of fake news connected to the business environment. Fake news typologies relating to the advancement of artificial intelligence new technologies are also explored. Findings & value added: By analysing important phrases, including false information, misinformation, disinformation, mal-information, and deepfakes, this research investigates the categorization of fake news linked to the business environment and social media concepts. It underscores the need for better truth comprehension and the significance of fact-checking in preventing the spread of false information, with governance and institutional implications in terms of the economics of artificial intelligence-generated fake news in the social media market. While previous studies have examined the fake news phenomenon from several angles, there is still a research gap, as the literature concentrates more on how fake news is consumed rather than how it is created. This research aims to bridge the gap by providing a comprehensive examination of fake news research from the perspectives of fake news typology, creation, detection, and regulatory means.

Fake news can have a significant negative impact on society because of the growing use of mobile devices and the worldwide increase in Internet access. It is therefore essential to develop a simple mathematical model to understand the online dissemination of fake news. In this study, we propose a point process model of the spread of fake news on Twitter. The proposed model describes the spread of a fake news item as a two-stage process: initially, fake news spreads as a piece of ordinary news; then, when most users start recognizing the falsity of the news item, that itself spreads as another news story. We validate this model using two datasets of fake news items spread on Twitter. We show that the proposed model is superior to the current state-of-the-art methods in accurately predicting the evolution of the spread of a fake news item. Moreover, a text analysis suggests that our model appropriately infers the correction time, i.e., the moment when Twitter users start realizing the falsity of the news item. The proposed model contributes to understanding the dynamics of the spread of fake news on social media. Its ability to extract a compact representation of the spreading pattern could be useful in the detection and mitigation of fake news.

The aim of this study was to evaluate the search for health information on the Internet and to determine the frequency and main means of spreading fake health news. A descriptive cross-sectional study was conducted through the virtual distribution of questionnaires on social media platforms in 2019 by using the snowball technique. The questionnaire collected information regarding sociodemographics, means used to clarify doubts about health, implementation of information obtained through the Internet, receipt of fake news, and means of transmission of fake news. Quantitative variables are described as means and standard deviations, and categorical variables are described as frequencies and percentages. The chi-square and Fisher's exact tests were used. Out of 1,195 respondents, 53% had followed Internet guidance without consulting a health professional, especially young people and individuals with low education levels (p<0.05). The resources most used to answer questions about health were a physician (78%) and Google (51%), and searches using the latter were more predominant among younger age groups (p<0.05). A large part of the sample (89.4%) had received fake news, and the main means of receipt were Facebook and WhatsApp. The Internet was the second most commonly used means to search for health information. A significant portion of the population adopts actions based on this information. The frequency of broadcasting fake news through this digital medium is high.

In perceptual decision-making, it is well-known that low confidence in a decision promotes information seeking, which increases subsequent decision accuracy. It is unclear, however, whether these mechanisms generalize to more complex domains such as fake news detection. Here, we tested whether confidence also drives information seeking in the context of fake news, whether seeking information improves fake news detection and sharing intentions, and whether we find confidence-driven confirmation biases. In an online study, 314 US citizens indicated whether news headlines were true or fake and could request additional information before finalizing their decision. First, in line with findings from perceptual decision-making, confidence was the main driver of information seeking for both fake and true news. Yet critically, whereas confidence lawfully scaled with decision accuracy for true news items, we found a U-shaped relationship for fake news items: decision accuracy was lowest for intermediate confidence, indicating that for fake headlines confidence is not a reliable signal of headline veracity. Second, while information seeking increased decision accuracy for true news items, for fake news items decision accuracy remained at chance level even after information seeking. Sharing intentions for fake news remained equally unaffected by information seeking, and mostly increased for belief-congruent true items. Finally, we found a confidence-driven confirmation bias for beliefs, but not for sharing intentions. Together, our results clarify that while confidence-driven information seeking adaptively increases decision accuracy for true news items, it is entirely ineffective in the context of fake news.