Fake News Stories Research Articles

Features Extraction Based on Probability Weighting for Fake News Classification on Social Media

Fake news is a massive problem globally, especially on social media. Most people spend a lot of time consuming social media every day, and it is very possible for people as social media users to receive fake news without realizing it. Primarily due to this situation, we developed a machine learning tool to detect fake news that operates with the aid of various algorithms such as Decision Tree, K-Nearest Neighbor, and Naïve Bayes. Our experiement is tested based on machine learning that selected only one technique used to classify the data by finding the model set. In addition, the performance of the set describes the classification of the model and the inconsistency solution for each iteration. This study proposed a model which used the probability weighting of the model in features extraction processing for data classification. The concept is the enhancement of probability weighting features that converge exactly the class labels of classification. Our work was also implemented based on traditional Count Vectorizer and TF-IDF Vectorizer sentiment analysis and combined probability weighting features for fake news articles. The experimental results of the work illustrate that the best accuracy achieved by a proposed model used probability weighting features to find out the impact of classifiers models. In addition, the results of experimental information is represented by enhancing the overall performance of Decision Tree, K-Nearest Neighbor, and Naïve Bayes with various datasets. In addition, the measures of precision, recall, F1-measure, AUC, and accuracy for each class and deep in each class were achieved and reached the highest performance of the proposed model.

Robust Detection of Fake News Using LSTM and GloVe Embeddings

The pervasive issue of fake news on digital platforms poses a significant threat to public opinion and trust in media. This paper addresses the problem of fake news detection by leveraging advanced natural language processing (NLP) techniques and deep learning models. Utilizing the ISOT Fake News Dataset, which comprises balanced samples of verified and fake news articles, we develop and evaluate two primary models: a Long Short-Term Memory (LSTM) network and a Convolutional Neural Network (CNN). The LSTM model employs pre-trained GloVe embeddings, followed by LSTM layers and a fully connected layer for classification, while the CNN model incorporates convolutional layers, max-pooling, and dropout for comparative analysis. Extensive pre-processing and exploratory data analysis (EDA) were conducted to clean the data and understand its characteristics. Our results demonstrate that the LSTM model outperforms the CNN model, achieving an accuracy of 99.58% on the test set. However, the high performance raises concerns about dataset biases, suggesting the need for more diverse and challenging datasets to ensure model robustness. Future work will focus on adversarial training and explainability techniques to enhance the model’s resilience and interpretability.

Fake News Incidents through the Lens of the DCAM Disinformation Blueprint

The emergence of the Internet and web technologies has magnified the occurrence of disinformation events and the dissemination of online fake news items. Fake news is a phenomenon where fake news stories are created and propagated online. Such events occur with ever increasing frequency, they reach a wide audience, and they can have serious real-life consequences. As a result, disinformation events are raising critical public interest concerns as in many cases online news stories of fake and disturbing events have been perceived as being truthful. However, even at a conceptual level, there is not a comprehensive approach to what constitutes fake news with regard to the further classification of individual occurrences and the detection/mitigation of actions. This work identifies the emergent properties and entities involved in fake news incidents and constructs a disinformation blueprint (DCAM-DB) based on cybercrime incident architecture. To construct the DCAM-DB in an articulate manner, the authors present an overview of the properties and entities involved in fake news and disinformation events based on the relevant literature and identify the most prevalent challenges. This work aspires to enable system implementations towards the detection, classification, assessment, and mitigation of disinformation events and to provide a foundation for further quantitative and longitudinal research on detection strategies.

K.Marx-Engels and Ho Chi Minh Viewpoints on Journalism - and Two Fake News Publishing Cases of Thanhnien.vn and Tuoitre.vn (Online Magazines) in Vietnam and Lessons from Indonesia, Japan Approaches

In this paper, by using qualitative analytical analysis with 2 case examples of Thanh nien and Tuoi tre newspapers (online) in Vietnam, in which there are history of publishing fakes news online from 2015, 2016, 2018, 2022 (with very bad editors Nguyen Ngoc Toan and Dang Thi Phuong Thao), as well as giaoduc.edu.vn and vietnamnet.vn in 2022 so we will address some points in this study based on answers for question: “What are regulatory lessons from Indonesia and Japan approaches on publishing fake news?”. We would suggest that there are penalties for negative behaviors of posting fake news online (any fake information) in the context of covid 19 epidemic. Tapsell (2019) defined ‘hoax news’ as similar to the more globally recognized term ‘fake news’: material deliberately fabricated and masqueraded as truth. At last, we will draw some lessons from K.Marx and Ho Chi Minh viewpoints on journalism for educating young generation in emerging markets such as Vietnam.

PREVENTION OF ONLINE FAKE NEWS ON SOCIAL MEDIA DURING COVID-19 PANDEMIC: A LITERATURE REVIEW APPROACH

The title of this paper is prevention of online fake news on social media during COVID-19 Pandemic: A literature review approach.Fake news or hoax has disrupted the original or authenticity of information, especially when social media become a flatform in disseminating fake news. And during COVID-19 Pandemic, online fake news on social media has been spread everywhere.The aims of this paper to discuss and explorer prevention of online fake news on social media during COVID-19 Pandemic. This paper applied qualitative research methods by using conceptual research and conceptual paper basically will provide logic and argument based on the literature review to explain a certain issue related to prevention of online fake news on social media during COVID-19 Pandemic.Based on study finding, prevention of online fake news on social media during COVID-19 Pandemic has clearly discuss from many sources. Bahri and Sosial, said there is two ways on prevention user from online fake news when using social media is with cognitive skill and critical thinking. Another way is stated by BNPT, social media user is suggested no to easily trust one piece of information when the source is not clear. Check news content and sources deliver news. Third and fourth, corrects and compare the information can trust or not. Fifth, do not spread it to other if user not sure validity of information. KOMINFO also said social media user must be careful with provocative titles, pay attention to the site address that information, fact check, check the authenticity, and joint anti Hoax forum to prevent social media user trapped in online fake news. Kementrian Agama also said, social media user must be able to filter the truth of the information circulating before it spreads. Second, remember ethics in social media. Third, be careful of unknown accounts. Fourth, do not upload information that conflicts with Sara. Fifth, use social media to expand the network. Sixth, include the source of the information content uploaded. Seventh, do not upload information when source is not clear. And lastly, using social media for self-development. Allah SWT also says in Surah Al-Hujarat:6 and An-Nur: 19 to verify all information that we get and do not spread fake news.

Automated Fake News Detection by LSTM Enabled with Optimal Feature Selection

Fake news plays a major role by broadcasting misinformation, which influences people’s knowledge or perceptions and distorts their decision-making and awareness. Online forums and social media have stimulated the broadcast of fake news by embedding it with truthful information. Thus, fake news has evolved into the main challenge of better impact in the information-driven community for intense fakesters. The detection of fake news articles that is generally found by considering the quality of the information in their news feeds under uncertain authenticity calls for automated tools. However, designing such tools is a major problem because of the multiple faces of fakesters. This paper offers a new text-analytics-driven method for detecting fake news to reduce the risks impacted by the consumption of fake news. The methodology for improved fake news detection focusses on four phases: (a) pre-processing, (b) feature extraction, (c) optimal feature selection and (d) classification. The pre-processing of the text data will be initially done by stop word removal, blank space removal and stemming. Further, the feature extraction is performed by term frequency-inverse document frequency, and grammatical analysis is done using mean, Q25, Q50, Q75, Max, Min and standard deviation. Then, the optimal feature selection is developed, which minimises the number of input variables. It is intended to reduce the number of input variables to improve the model’s performance by minimising the computational cost of modelling. An improved meta-heuristic algorithm called successive position-based barnacles mating optimisation is used for optimal feature selection and classification. As the main contribution, the influence of deep learning is employed, which employs optimised long short-term memory. Finally, the result shows the superiority in terms of different significant measures by the proposed model over other methods for fake news detection experimentally done on a publicly available benchmark dataset.

Digital misinformation and fake news detection using WoT integration with Asian social networks fusion based feature extraction with text and image classification by machine learning architectures

Fake News contains potentially false data that is verified. Concerns about author accountability are being addressed by organizations such as the House of Commons and the Crosscheck initiative. So here proposed system designs the web classification based on ontology with fuzzy logic. The feature extraction has been carried out using deep learning techniques. The initial feature extraction has been done using AE-CNN (Autoencoder with convolution neural network). After the characteristics were extracted, high dimensionality indexing was used. Then based on indexing measures ranking process takes place to identify the feature whether it is image or text. After this training part the trained file has been annotated with the fuzzy-based ontology rules to detect online fake news web content. Then the decision is taken and finally genetic algorithm-based deep learning is carried out for web classification. The simulation results have been analyzed regarding accuracy, precision, recall and F-1 score.

‘Abba Kyari did not die of coronavirus’: Social media and fake news during a global pandemic in Nigeria

This study examined the influence of fake news online on how social media users viewed and reacted to the COVID-19 pandemic in Nigeria. Analyses of an online survey (N = 254) and contents from Twitter users in Nigeria from the hashtags: ‘#coronavirusNigeria’ and ‘#covid19Nigeria’ (N = 10,408), reveal that social media users in Nigeria used Twitter to inform and educate Twitter users as well as debunking fake news stories about the virus to prevent purveyors of fake news from misleading Twitter users in Nigeria. Findings further indicate that those who use social media platforms and national television as main sources of news and are less educated, are statistically more likely to believe fake news about the virus than those who are educated and used newspapers as main sources of news. Consequently, the study recommends that Nigerian political leaders enact policies that they can observe as their inability to adhere to their own lockdowns powered fake news about COVID-19 in Nigeria.

A benchmark study of machine learning models for online fake news detection

A benchmark study of machine learning models for online fake news detection

Modeling the influencing factors of electronic word-of-mouth about CSR on social networking sites.

Corporate social responsibility (CSR) information can now be disseminated via social networking sites. An organization's image and upcoming portfolios are directly affected by electronic word of mouth (eWOM). It generates from its customers, employees, and other stakeholders. We developed a critical model to enlighten the behavior to share and comment on a negative news story about CSR displayed on Wechat and QQ to figure out what was causing this behavior. Structural equation modeling (SEM) and the partial least squares regression (PLS) approach were used to conduct a self-administered survey of hotel customers in China. Social and environmental awareness, information usefulness, corporate image, and a company's motivation to comment and share on CSR news were all considered explanatory variables in our study. We asked 300 Wechat and QQ users to rate a fake environmental news story. We found that social and environmental awareness affects the effectiveness of information and the attitude toward behavior, which may describe the eWOM intent of the particular news. On the other hand, corporate reputation could discourage people from disseminating eWOM and sharing the news with their social linkage contacts. The findings of the study suggest having a better understanding of how specific CSR activities can increase customers' commitment, which leads to positive eWOM, will benefit the hotel industry.

Publishing Fake Information Online-Case of Online Vietnam Magazines (Thanhnien Newspaper, Tuoi tre newspaper, Vietnamnet.vn, dantri.com.vn, giaoduc.edu.vn, sctv.com.vn, etc.) From an Approach of German and EU Laws and Cybersecurity Regulations

Publishing fake information have at least three bad effects on the community: creating disinformation, anxiety, and disorder in society. Still using qualitative analytical methods with synthesis and inductive methods, the authors will address 2 cases of Vietnam magazines: thanhnien. vn and tuoitre.vn, vietnamnet. vn, giaoduc.edu.vn, dantri.com.vn, (online newspapers) and recently sctv.com.vn with their issue of publishing fake news online, which increasing as a problem in recent years 2015-2020. In this paper, we also use the European approach and laws on exploring the issue of publishing and delivering false information via the internet and social media. Last but not least, the views and ideologies of V.I Lenin and Ho Chi Minh on journalism and journalists are mentioned for educating the young generation.

A digital media literacy intervention for older adults improves resilience to fake news

Older adults are especially susceptible to fake news online, possibly because they are less digitally literate compared to younger individuals. Interventions for older adults have emerged to improve digital literacy, although there has been little evaluation of their effectiveness in improving older adults’ resilience to fake news. We report the results of a digital literacy intervention for older adults administered during the 2020 U.S. election. The intervention was a 1-hour, self-directed series of interactive modules designed to teach concepts and skills for identifying misinformation online. Consistent with our pre-registered hypothesis, older adults (Mage = 67) in the treatment condition (N = 143) significantly improved their likelihood of accurately discerning fake from true news from 64% pre-intervention to 85% post-intervention. In contrast, older adults in the control condition (N = 238) did not significantly improve (from 55% to 57%). The treated older adults were also more likely to employ strategies for identifying misinformation online compared to pre-intervention and the control group.

Deep Convolutional Neural Network for Fake News Detection over Online Social Networks

With the proliferation of Online Social Networks (OSNs), people of all walks of life are able to have instant sharing of information and opinions. Thus they became virtual platforms for exchanging views and making friends. The news shared over social media helps in understanding various aspects. They also come in the form of reviews or product or service news that will help people in making well informed decisions. As people are increasingly relying on social media to ascertain knowledge of products or services, the news over social media needs to be genuine. Unfortunately, it is observed that there are some fake news articles or reviews over OSNs. Unless, such news is identified and eliminated, it will cause people in making wrong decisions. They influence of fake news on people will defeater the purpose of social media content. The existing machine learning algorithms used for fake news detection provide a great deal of advantages in detection of fake news. However, they suffer from deterioration of performance due to lack of multiple layers of training. This is achieved by deep learning techniques. In this paper, a deep Convolutional Neural Network (CNN) is used for prediction of fake news. An algorithm known as Deep Fake News Detection (DFND) is proposed and implemented. A prototype application is built using Python data science platform to show the utility of the proposed system. The empirical results of the proposed system are compared with that of existing ones. Keywords – Deep Learning, Convolutional Neural Network, Fake News Detection Deep Fake News Detection Algorithm

Fake news on the internet: a literature review, synthesis and directions for future research

PurposeThe extensive distribution of fake news on the internet (FNI) has significantly affected many lives. Although numerous studies have recently been conducted on this topic, few have helped us to systematically understand the antecedents and consequences of FNI. This study contributes to the understanding of FNI and guides future research.Design/methodology/approachDrawing on the input–process–output framework, this study reviews 202 relevant articles to examine the extent to which the antecedents and consequences of FNI have been investigated. It proposes a conceptual framework and poses future research questions.FindingsFirst, it examines the “what”, “why”, “who”, “when”, “where” and “how” of creating FNI. Second, it analyses the spread features of FNI and the factors that affect the spread of FNI. Third, it investigates the consequences of FNI in the political, social, scientific, health, business, media and journalism fields.Originality/valueThe extant reviews on FNI mainly focus on the interventions or detection of FNI, and a few analyse the antecedents and consequences of FNI in specific fields. This study helps readers to synthetically understand the antecedents and consequences of FNI in all fields. This study is among the first to summarise the conceptual framework for FNI research, including the basic relevant theoretical foundations, research methodologies and public datasets.

Multi-scale Machine Learning Prediction of the Spread of Arabic Online Fake News

There are a lot of research studies that look at "fake news" from an Arabic online source, but they don't look at what makes those fake news spread. The threat grows, and at some point, it gets out of hand. That's why this paper is trying to figure out how to predict the features that make Arabic online fake news spread. It's using Naive Bayes, Logistic Regression, and Random forest of Machine Learning to do this. Online news stories that were made up were used. They are found by using Term Frequency-Inverse Document Frequency (TF-IDF). The best partition for testing and validating the prediction was chosen at random and used in the analysis. So, all three machine learning classifications for predicting fake news in Arabic online were done. The results of the experiment show that Random Forest Classifier outperformed the other two algorithms. It had the best TF-IDF with an accuracy of 86 percent. Naive Bayes had an accuracy rate of 84%, and Logistic Regression had an accuracy rate of 85%, so they all did well. As such, the model shows that the features in TF-IDF are the most essential point about the content of an online Arabic fake news.

Trust, Media Credibility, Social Ties, and the Intention to Share towards Information Verification in an Age of Fake News.

Social media is now the primary form of communication between internet users and has soared in popularity, which has directly impacted the spread of the phenomenon of fake news. Fake news is not only a widespread phenomenon; it is also problematic and dangerous for society. The aim of this study is to understand the phenomenon of fake news better. The study utilised a structural modelling equation in order to identify how Polish society perceives the problem of fake news and assess the extent to which it trusts content that is published on the internet. The key goal was to determine what factors have the most significant influence on the verification of information being viewed on the internet. By deploying the partial least squares method of validation, SmartPLS3 software was used to process the survey results. The strongest positive effect on information verification behaviour was found to be fake news awareness, which was followed by the intention to share information. The research did not consider any clear connections that may exist between the nature of fake news and its recipient; however, much of the fake news that appears on the internet is political in nature. The study can be used by news reporting companies and provides preliminary information for developers responsible for running social media sites as well as users who want to combat and limit the spread of fake news online. This study expands on the available literature related to fake news by identifying the effects on information verification behaviour of fake news awareness and the intention to share data.

Arabic Fake News Detection Based on Textual Analysis

Over the years, social media has had a considerable impact on the way we share information and send messages. With this comes the problem of the rapid distribution of fake news which can have negative impacts on both individuals and society. Given the potential negative influence, detecting unmonitored ‘fake news’ has become a critical issue in mainstream media. While there are recent studies that built machine learning models that detect fake news in several languages, lack of studies in detecting fake news in the Arabic language is scare. Hence, in this paper, we study the issue of fake news detection in the Arabic language based on textual analysis. In an attempt to address the challenges of authenticating news, we introduce a supervised machine learning model that classifies Arabic news articles based on their context’s credibility. We also introduce the first dataset of Arabic fake news articles composed through crowdsourcing. Subsequently, to extract textual features from the articles, we create a unique approach of forming Arabic lexical wordlists and design an Arabic Natural Language Processing tool to perform textual features extraction. The findings of this study promises great results and outperformed human performance in the same task.

Quantifying the impacts of online fake news on the equity value of social media platforms – Evidence from Twitter

Online Fake News characterized by falsehood and ambiguity is significantly shaking up various aspects of social, economic, and political life across the globe. In addition, it can also be detrimental to the existence of social media platforms. In this research, by synthesizing the prior literature on negativity bias in online settings, reasons for platform failure and the characteristics of social media platforms, we explain the mechanisms through which online fake news impacts the equity value of these platforms. We also develop and implement a two-stage Bayesian Vector Autoregression technique to test these mechanisms using a dataset from Twitter. Our results show that falsehood results in an equity value loss of approximately 2.11 Million USD over a ten-day period while a mere 67.17 Million falsehood tweets can interact with ambiguity to produce a loss of 10 Million USD. We also find that ambiguity helps mitigate the negative impact of fake news which we attribute to the fact that users converse and disambiguate news that arrives on social media platforms. Our research has theoretical and practical implications for the impact of fake news on these platforms and their valuations.

Faking and Conspiring about COVID-19: A Discursive Approach

In the more general climate of post-truth - a social trend reflecting a disregard for reliable ways of knowing what is true, mostly acted through massive use of misinformation and rhetoric calling for emotions - an alarming “infodemic” accompanied the COVID-19 pandemic, affecting healthy attitudes and behaviors and further lessening trust in science, institutions, and traditional media. Its two main representative items, fake and conspiracy news, have been widely analyzed in psycho-social research, even if scholars mostly acknowledged the cognitive and social dimensions of those items and devoted less attention to their discursive construction. In addition, these works did not directly compare and differentiate fake and conspiracy pathways. In order to address this gap and promote a wider understanding of these matters, a qualitative investigation of an Italian sample of 112 fake and conspiracy news articles, mostly spread during the first two COVID-19 “waves” (from March 2020 to January 2021) was realized. Our sample gathered news specifically coming from social media posts, representing easy and fast channels for viral content diffusion. We analyzed the selected texts by means of Diatextual Analysis and Discursive Action Model models, aimed to (a) offer “in depth” fine-grained analysis of the psycholinguistic and argumentative features of fake and conspiracy news, and (b) differentiate them in line with the classical Aristotle’s rhetoric stances of logos, ethos, and pathos, thus bridging traditional and current lines of thinking. Even though they may share common roots set in the post-truth climate, fake and conspiracy news engage in different rhetoric patterns since they present different enjeu and construct specific epistemic pathways. Implications for health- and digital-literacy are debated.

Health-related fake news during the COVID-19 pandemic: perceived trust and information search

PurposeHealth-related online fake news (HOFN) has become a major social problem. HOFN can lead to the spread of ineffective and even harmful remedies. The study aims to understand Internet users' responses to HOFN during the coronavirus (COVID-19) pandemic using the protective action decision model (PADM).Design/methodology/approachThe authors collected pandemic severity data (regional number of confirmed cases) from government websites of the USA and China (Studies 1 and 2), search behavior from Google and Baidu search engines (Studies 1 and 2) and data regarding trust in two online fake news stories from two national surveys (Studies 2 and 3). All data were analyzed using a multi-level linear model.FindingsThe research detected negative time-lagged relationships between pandemic severity and regional HOFN search behavior by three actual fake news stories from the USA and China (Study 1). Importantly, trust in HOFN served as a mediator in the time-lagged relationship between pandemic severity and search behavior (Study 2). Additionally, the relationship between pandemic severity and trust in HOFN varied according to individuals' perceived control (Study 3).Originality/valueThe authors' results underscore the important role of PADM in understanding Internet users' trust in and search for HOFN. When people trust HOFN, they may seek more information to implement further protective actions. Importantly, it appears that trust in HOFN varies with environmental cues (regional pandemic severity) and with individuals' perceived control, providing insight into developing coping strategies during a pandemic.