A Deep Learning Based Approach for Fake News Detection

2021 ◽  
pp. 388-394
Author(s):  
Srishti Sharma ◽  
Vaishali Kalra
Keyword(s):  
Social Networks ◽  
Social Media ◽  
Language Processing ◽  
Research Effort ◽  
Political Polarization ◽  
Fake News ◽  
User Characteristics ◽  
Mainstream Media ◽  
Boosting Method ◽  
Partisan Conflict

Owing to the rapid explosion of social media platforms in the past decade, we spread and consume information via the internet at an expeditious rate. It has caused an alarming proliferation of fake news on social networks. The global nature of social networks has facilitated international blowout of fake news. Fake news has proven to increase political polarization and partisan conflict. Fake news is also found to be more rampant on social media than mainstream media. The evil of fake news is garnering a lot of attention and research effort. In this work, we have tried to handle the spread of fake news via tweets. We have performed fake news classification by employing user characteristics as well as tweet text. Thus, trying to provide a holistic solution for fake news detection. For classifying user characteristics, we have used the XGBoost algorithm which is an ensemble of decision trees utilising the boosting method. Further to correctly classify the tweet text we used various natural language processing techniques to preprocess the tweets and then applied a sequential neural network and state-of-the-art BERT transformer to classify the tweets. The models have then been evaluated and compared with various baseline models to show that our approach effectively tackles this problemOwing to the rapid explosion of social media platforms in the past decade, we spread and consume information via the internet at an expeditious rate. It has caused an alarming proliferation of fake news on social networks. The global nature of social networks has facilitated international blowout of fake news. Fake news has proven to increase political polarization and partisan conflict. Fake news is also found to be more rampant on social media than mainstream media. The evil of fake news is garnering a lot of attention and research effort. In this work, we have tried to handle the spread of fake news via tweets. We have performed fake news classification by employing user characteristics as well as tweet text. Thus, trying to provide a holistic solution for fake news detection. For classifying user characteristics, we have used the XGBoost algorithm which is an ensemble of decision trees utilising the boosting method. Further to correctly classify the tweet text we used various natural language processing techniques to preprocess the tweets and then applied a sequential neural network and state-of-the-art BERT transformer to classify the tweets. The models have then been evaluated and compared with various baseline models to show that our approach effectively tackles this problem

scholarly journals Intelligent Detection of False Information in Arabic Tweets Utilizing Hybrid Harris Hawks Based Feature Selection and Machine Learning Models

Symmetry ◽  
2021 ◽  
Vol 13 (4) ◽  
pp. 556
Author(s):  
Thaer Thaher ◽  
Mahmoud Saheb ◽  
Hamza Turabieh ◽  
Hamouda Chantar
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Feature Selection ◽  
Language Processing ◽  
User Profile ◽  
Vital Role ◽  
Classification Model ◽  
Fake News ◽  
False Information ◽  
Social Media Platforms

Fake or false information on social media platforms is a significant challenge that leads to deliberately misleading users due to the inclusion of rumors, propaganda, or deceptive information about a person, organization, or service. Twitter is one of the most widely used social media platforms, especially in the Arab region, where the number of users is steadily increasing, accompanied by an increase in the rate of fake news. This drew the attention of researchers to provide a safe online environment free of misleading information. This paper aims to propose a smart classification model for the early detection of fake news in Arabic tweets utilizing Natural Language Processing (NLP) techniques, Machine Learning (ML) models, and Harris Hawks Optimizer (HHO) as a wrapper-based feature selection approach. Arabic Twitter corpus composed of 1862 previously annotated tweets was utilized by this research to assess the efficiency of the proposed model. The Bag of Words (BoW) model is utilized using different term-weighting schemes for feature extraction. Eight well-known learning algorithms are investigated with varying combinations of features, including user-profile, content-based, and words-features. Reported results showed that the Logistic Regression (LR) with Term Frequency-Inverse Document Frequency (TF-IDF) model scores the best rank. Moreover, feature selection based on the binary HHO algorithm plays a vital role in reducing dimensionality, thereby enhancing the learning model’s performance for fake news detection. Interestingly, the proposed BHHO-LR model can yield a better enhancement of 5% compared with previous works on the same dataset.

What the fake? Assessing the extent of networked political spamming and bots in the propagation of #fakenews on Twitter

2019 ◽  
Vol 43 (1) ◽  
pp. 53-71 ◽  
Author(s):  
Ahmed Al-Rawi ◽  
Jacob Groshek ◽  
Li Zhang
Keyword(s):  
Social Media ◽  
Data Sets ◽  
Fake News ◽  
Content Type ◽  
Mainstream Media ◽  
News Discourse ◽  
Twitter Data ◽  
News Organizations ◽  
Twitter Users ◽  
One Year

PurposeThe purpose of this paper is to examine one of the largest data sets on the hashtag use of #fakenews that comprises over 14m tweets sent by more than 2.4m users.Design/methodology/approachTweets referencing the hashtag (#fakenews) were collected for a period of over one year from January 3 to May 7 of 2018. Bot detection tools were employed, and the most retweeted posts, most mentions and most hashtags as well as the top 50 most active users in terms of the frequency of their tweets were analyzed.FindingsThe majority of the top 50 Twitter users are more likely to be automated bots, while certain users’ posts like that are sent by President Donald Trump dominate the most retweeted posts that always associate mainstream media with fake news. The most used words and hashtags show that major news organizations are frequently referenced with a focus on CNN that is often mentioned in negative ways.Research limitations/implicationsThe research study is limited to the examination of Twitter data, while ethnographic methods like interviews or surveys are further needed to complement these findings. Though the data reported here do not prove direct effects, the implications of the research provide a vital framework for assessing and diagnosing the networked spammers and main actors that have been pivotal in shaping discourses around fake news on social media. These discourses, which are sometimes assisted by bots, can create a potential influence on audiences and their trust in mainstream media and understanding of what fake news is.Originality/valueThis paper offers results on one of the first empirical research studies on the propagation of fake news discourse on social media by shedding light on the most active Twitter users who discuss and mention the term “#fakenews” in connection to other news organizations, parties and related figures.

scholarly journals An Analysis of Fake Narratives on Social Media during 2019 Indonesian Presidential Election

2020 ◽  
Vol 36 (4) ◽  
pp. 351-368
Author(s):  
Vience Mutiara Rumata ◽  
◽  
Fajar Kuala Nugraha ◽  
Keyword(s):  
Social Media ◽  
Presidential Election ◽  
Political Polarization ◽  
Political Discussion ◽  
Fake News ◽  
Data Set ◽  
Political Situation ◽  
Radical Group ◽  
Time Frames ◽  
Election Studies

Social media become a public sphere for political discussion in the world, with no exception in Indonesia. Social media have broadened public engagement but at the same time, it creates an inevitable effect of polarization particularly during the heightened political situation such as a presidential election. Studies found that there is a correlation between fake news and political polarization. In this paper, we identify and the pattern of fake narratives in Indonesia in three different time frames: (1) the Presidential campaign (23 September 2018 -13 April 2019); (2) the vote (14-17 April 2019); (3) the announcement (21-22 May 2019). We extracted and analyzed a data-set consisting of 806,742 Twitter messages, 143 Facebook posts, and 16,082 Instagram posts. We classified 43 fake narratives where Twitter was the most used platform to distribute fake narratives massively. The accusation of Muslim radical group behind Prabowo and Communist accusation towards the incumbent President Joko Widodo were the two top fake narratives during the campaign on Twitter and Facebook. The distribution of fake narratives to Prabowo was larger than that to Joko Widodo on those three platforms in this period. On the contrary, the distribution of fake narratives to Joko Widodo was significantly larger than that to Prabowo during the election and the announcement periods. The death threat of Joko Widodo was top fake narratives on these three platforms. Keywords: Fake narratives, Indonesian presidential election, social media, political polarization, post.

scholarly journals Fake and Real News detection Using Python

2020 ◽  
pp. 423-428
Author(s):  
Fakhra Akhtar ◽  
Faizan Ahmed Khan
Keyword(s):  
Social Media ◽  
Fake News ◽  
False Information ◽  
False Negatives ◽  
Mainstream Media ◽  
Social Media Platforms ◽  
The Masses ◽  
Negative Attributes ◽  
Entire Dataset ◽  
F Measure

<p>In the digital age, fake news has become a well-known phenomenon. The spread of false evidence is often used to confuse mainstream media and political opponents, and can lead to social media wars, hatred arguments and debates.Fake news is blurring the distinction between real and false information, and is often spread on social media resulting in negative views and opinions. Earlier Research describe the fact that false propaganda is used to create false stories on mainstream media in order to cause a revolt and tension among the masses The digital rights foundation DRF report, which builds on the experiences of 152 journalists and activists in Pakistan, presents that more than 88 % of the participants find social media platforms as the worst source for information, with Facebook being the absolute worst. The dataset used in this paper relates to Real and fake news detection. The objective of this paper is to determine the Accuracy , precision , of the entire dataset .The results are visualized in the form of graphs and the analysis was done using python. The results showed the fact that the dataset holds 95% of the accuracy. The number of actual predicted cases were 296. Results of this paper reveals that The accuracy of the model dataset is 95.26 % the precision results 95.79 % whereas recall and F-Measure shows 94.56% and 95.17% accuracy respectively.Whereas in predicted models there are 296 positive attributes , 308 negative attributes 17 false positives and 13 false negatives. This research recommends that authenticity of news should be analysed first instead of drafting an opinion, sharing fake news or false information is considered unethical journalists and news consumers both should act responsibly while sharing any news.</p>

Fake News Polarization for Sentiment Analysis

2022 ◽  
pp. 255-263
Author(s):  
Chirag Visani ◽  
Vishal Sorathiya ◽  
Sunil Lavadiya
Keyword(s):  
Social Networks ◽  
Social Media ◽  
Sentiment Analysis ◽  
Online Social Networks ◽  
Opinion Mining ◽  
Easy Access ◽  
The Internet ◽  
Fake News ◽  
Social Media Networks ◽  
Sentiment Mining

The popularity of the internet has increased the use of e-commerce websites and news channels. Fake news has been around for many years, and with the arrival of social media and modern-day news at its peak, easy access to e-platform and exponential growth of the knowledge available on social media networks has made it intricate to differentiate between right and wrong information, which has caused large effects on the offline society already. A crucial goal in improving the trustworthiness of data in online social networks is to spot fake news so the detection of spam news becomes important. For sentiment mining, the authors specialise in leveraging Facebook, Twitter, and Whatsapp, the most prominent microblogging platforms. They illustrate how to assemble a corpus automatically for sentiment analysis and opinion mining. They create a sentiment classifier using the corpus that can classify between fake, real, and neutral opinions in a document.

scholarly journals Simulacra in the Age of Social Media: Baudrillard as the Prophet of Fake News

2020 ◽  
pp. 019685992097715
Author(s):  
James Morris
Keyword(s):  
Social Media ◽  
General Trend ◽  
Digital Age ◽  
The United States ◽  
The Political ◽  
Fake News ◽  
Donald Trump ◽  
Online Technology ◽  
Mainstream Media ◽  
Mass Reproduction

“Fake News” has been a frequent topic in the last couple of years. The phenomenon has particularly been cited with regards to the election of Donald Trump to the presidency of the United States. The creation of “post truth” reports that are disseminated via the Web and social media has been treated as something new, a product of the digital age, and a reason to be concerned about the effects of online technology. However, this paper argues that fake news should be considered as part of a continuum with forms of media that went before in the 20th Century, and the general trend of postmodernity detailed by Baudrillard. The simulation of communications media and mass reproduction was already evident and has merely progressed in the digital age rather than the latter providing a wholly new context. The paper concludes by asking whether the political havoc caused by fake news has an antidote, when it appears to be a by-product of media simulacra’s inherent lack of connection to the real. In a communications landscape where the misrepresentations of the so-called “Mainstream Media” are decried using even more questionable “memes” on social media, is there any possibility for truth?

scholarly journals Exploring Adversarial Attacks and Defences for Fake Twitter Account Detection

Technologies ◽  
2020 ◽  
Vol 8 (4) ◽  
pp. 64
Author(s):  
Panagiotis Kantartopoulos ◽  
Nikolaos Pitropakis ◽  
Alexios Mylonas ◽  
Nicolas Kylilis
Keyword(s):  
Machine Learning ◽  
Social Networks ◽  
Social Media ◽  
Test Phase ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Fake News ◽  
Specific Subject ◽  
Learning Techniques ◽  
Twitter Account

Social media has become very popular and important in people’s lives, as personal ideas, beliefs and opinions are expressed and shared through them. Unfortunately, social networks, and specifically Twitter, suffer from massive existence and perpetual creation of fake users. Their goal is to deceive other users employing various methods, or even create a stream of fake news and opinions in order to influence an idea upon a specific subject, thus impairing the platform’s integrity. As such, machine learning techniques have been widely used in social networks to address this type of threat by automatically identifying fake accounts. Nonetheless, threat actors update their arsenal and launch a range of sophisticated attacks to undermine this detection procedure, either during the training or test phase, rendering machine learning algorithms vulnerable to adversarial attacks. Our work examines the propagation of adversarial attacks in machine learning based detection for fake Twitter accounts, which is based on AdaBoost. Moreover, we propose and evaluate the use of k-NN as a countermeasure to remedy the effects of the adversarial attacks that we have implemented.

scholarly journals New explainability method for BERT-based model in fake news detection

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Mateusz Szczepański ◽  
Marek Pawlicki ◽  
Rafał Kozik ◽  
Michał Choraś
Keyword(s):  
Artificial Intelligence ◽  
Social Media ◽  
Language Processing ◽  
High Performance ◽  
Real Life ◽  
Contemporary Society ◽  
Fake News ◽  
Interpretable Model ◽  
Deep Integration ◽  
The Impact

AbstractThe ubiquity of social media and their deep integration in the contemporary society has granted new ways to interact, exchange information, form groups, or earn money—all on a scale never seen before. Those possibilities paired with the widespread popularity contribute to the level of impact that social media display. Unfortunately, the benefits brought by them come at a cost. Social Media can be employed by various entities to spread disinformation—so called ‘Fake News’, either to make a profit or influence the behaviour of the society. To reduce the impact and spread of Fake News, a diverse array of countermeasures were devised. These include linguistic-based approaches, which often utilise Natural Language Processing (NLP) and Deep Learning (DL). However, as the latest advancements in the Artificial Intelligence (AI) domain show, the model’s high performance is no longer enough. The explainability of the system’s decision is equally crucial in real-life scenarios. Therefore, the objective of this paper is to present a novel explainability approach in BERT-based fake news detectors. This approach does not require extensive changes to the system and can be attached as an extension for operating detectors. For this purposes, two Explainable Artificial Intelligence (xAI) techniques, Local Interpretable Model-Agnostic Explanations (LIME) and Anchors, will be used and evaluated on fake news data, i.e., short pieces of text forming tweets or headlines. This focus of this paper is on the explainability approach for fake news detectors, as the detectors themselves were part of previous works of the authors.

scholarly journals Post-truth and trust in journalism: an analysis of credibility indicators in Brazilian venues

2019 ◽  
Vol 15 (3) ◽  
pp. 452-473
Author(s):  
Marcelo Träsel ◽  
Sílvia Lisboa ◽  
Giulia Reis Vinciprova
Keyword(s):  
Social Networks ◽  
Digital Networks ◽  
Fake News ◽  
Redes Sociales ◽  
Mainstream Media ◽  
News Stories ◽  
Current Scenario

The terms ‘fake news’ and ‘post-truth’ have been used to describe the augmented dissemination potential of misinformation in digital networks in the second decade of the years 2000. In Brazil, different actors have been exploiting digital social networks for political purposes, disseminating content that imitates legitimate journalistic material, often obtaining better audience metrics than the news stories published by mainstream media. This article is divided into two parts. First, defines the term pseudojournalism to classify fraudulent texts that use journalistic narrative resources to deceive the audience. Second, it presents the results of an analysis of 23 political content producers with the greatest audience on Facebook in Brazil, based on the credibility indicators developed by Projeto Credibilidade (Trust Project). The results suggest that, in the current scenario, it is not possible to distinguish the quality journalism from pseudojournalism based on the characteristics of the websites and articles published by political content producers.Os termos “notícias falsas” e “pós-verdade” vêm sendo usados para descrever a potencialização da desinformação nas redes digitais na segunda década dos anos 2000. No Brasil, diversos atores vêm instrumentalizando as redes sociais para disputas políticas, espalhando conteúdo falso que imita materiais jornalísticos legítimos, muitas vezes obtendo mais audiência do que o noticiário de veículos tradicionais. Este artigo se divide em duas partes. Na primeira, conceitua o termo pseudojornalismo para classificar textos fraudulentos que usam os recursos narrativos jornalísticos para ludibriar a audiência. Na segunda, apresenta os resultados de uma análise de 23 produtores de conteúdo político do país com maior audiência no Facebook, a partir dos indicadores de credibilidade desenvolvidos pelo Projeto Credibilidade (Trust Project). Os resultados sugerem que, no cenário atual, não é possível distinguir o jornalismo de qualidade do pseudojornalismo a partir das características dos websites e matérias publicadas por produtores de conteúdo político.Las expresiones “noticias falsas” y “posverdad” vienen siendo utilizados para describir la potencialización de la desinformación en las redes digitales en la segunda década de los años 2000. En Brasil, distintos actores vienen instrumentalizando las redes sociales para disputas políticas, diseminando contenido falso que simula materiales periodísticos legítimos, obteniendo, a menudo, mayor audiencia que el noticiero de medios tradicionales. Este artículo está dividido en dos partes. Primero, conceptualiza el término pseudoperiodismo para calificar textos fraudulentos que utilizan los recursos de narración típicos del periodismo para engañar a la audiencia. En segundo lugar, presenta los resultados de un análisis de 23 productores de contenido político del país con mayor audiencia en Facebook, a partir de los indicadores de credibilidad desarrollados por el Proyecto Credibilidad (Trust Project). Los resultados sugieren que, en el escenario actual, no es posible diferenciar el periodismo de calidad del pseudoperiodismo a partir de las características de los sitios web y de materias publicadas por productores de contenido político.

scholarly journals Anxiety disorder and consumption of social media in Brazil

2020 ◽  
Vol 8 (5) ◽  
pp. 316-326
Author(s):  
Tarcisio Torres Silva
Keyword(s):  
Social Networks ◽  
Social Media ◽  
Anxiety Disorder ◽  
World Health ◽  
Political Polarization ◽  
The Political ◽  
Social Changes ◽  
The World ◽  
Health Organization ◽  
The City

Brazilian population spends a lot of time on social media. The average access from any device is 3 hours and 39 minutes (The Global, 2018). On the other hand, the country leads the numbers of anxiety disorder among the population. According to the World Health Organization, the incidence in the country is 9.3%, while the world average is 3.5%. This number is even higher in big cities, reaching 19.9% in the city of São Paulo (Horta, 2019). Possible causes are economic instability, social changes and violence (Horta, 2019). Add to that the political polarization in recent years and the intensive use of gadgets, private chat applications, such as Whatsapp, and social networks. In this work, we focus on the influence of social networks in the development of Brazilian anxiety. Our hypothesis is that the intensity of use reinforces the existence of other factors of anxiety increase (economy, violence, political division, etc.) through the sharing of news, besides adding others, such as self-display, performativity and the need of always being in evidence in social networks. As a method, we will work with content analysis (news and images) from the main social networking platforms used in Brazil.

Sign in / Sign up

Export Citation Format

Share Document