Recognition Of News Research Articles

Federated Learning in the Detection of Fake News Using Deep Learning as a Basic Method.

This article explores the possibilities for federated learning with a deep learning method as a basic approach to train detection models for fake news recognition. Federated learning is the key issue in this research because this kind of learning makes machine learning more secure by training models on decentralized data at decentralized places, for example, at different IoT edges. The data are not transformed between decentralized places, which means that personally identifiable data are not shared. This could increase the security of data from sensors in intelligent houses and medical devices or data from various resources in online spaces. Each station edge could train a model separately on data obtained from its sensors and on data extracted from different sources. Consequently, the models trained on local data on local clients are aggregated at the central ending point. We have designed three different architectures for deep learning as a basis for use within federated learning. The detection models were based on embeddings, CNNs (convolutional neural networks), and LSTM (long short-term memory). The best results were achieved using more LSTM layers (F1 = 0.92). On the other hand, all three architectures achieved similar results. We also analyzed results obtained using federated learning and without it. As a result of the analysis, it was found that the use of federated learning, in which data were decomposed and divided into smaller local datasets, does not significantly reduce the accuracy of the models.

A Quasi Experiment on the Effectiveness of Social Media Literacy Skills Training for Combating Fake News Proliferation

This study experimented the effect of social media literacy skills training as an intervention strategy for combating fake news. We carried out a quasi-experiment in one public university in Nigeria and randomly divided 480 participants into equal parts to form a control and experimental group. The respondents in the experimental group were exposed to 8 weeks of training to improve their social media literacy skills to fight fake news. We realised that social media literacy skills training is effective in increasing social media knowledge, users’ recognition of fake news, their tendency to verify information, and reduce the inclination to share false news since users exposed to social media literacy training can differentiate between accurate and made-up news compared to their counterparts who were not exposed to the same training. Implications for research and practice were discussed.

Machine Learning for Fake News Detection Analysis

The COVID-19 outbreak has required some health and financial decisions to be made in an unwieldy manner. This has spread uncertainty and lies all over the world. The transmission of false information has been compounded by the problems with fake news. Manyof them gave up on newspapers, magazines, and other print media in favorof Internetpleasure. Online entertainment has become the primary news source for a sizable percentage of the population due to its ease of access, low cost, and rapid spread. In some circumstances, bogus information spreads faster than true information togain popularity over internet entertainment and divert people from the underlying issues. People spread false information using online entertainment for commercial and political benefit. To avoid a harmful influence on society, it is critical to immediately recognizebogus information in all systems. To demonstrate the efficiency of the grouping on the dataset, we produced and tested numerous AI computations independently for this assignment, which looks into research on the recognition of fake news. The Jupyter Notebookstage of this project was used, and the execution was assessed.

Recognizing fake news based on natural language processing using the BM25 algorithm with fine-tuned parameters

The object of the research is the method of natural language processing (NLP) with balanced parameters of the BestMatch25 (ВМ25) algorithm to recognize and classify fake news based on natural language processing (NLP). The unsatisfactory accuracy and speed of existing methods for detecting fake news in unstructured input data demanded the development of a new approach for their effective detection. The study investigated the BM25 algorithm, methods for selecting parameters k1 and b, and their impact on the algorithm's effectiveness in detecting fake news. It was established that precise and detailed adjustment of these parameters is crucial in achieving optimal accuracy and data processing speed. The results showed that the successful selection of BM25 parameters improves the model's accuracy by up to 14 % compared to standard term frequency – inverse document frequency (TF-IDF) calculations. These results were made possible by experimentally tuning different combinations of k1 and b parameters, in which the algorithm shows the best speed indicator or the most accurate estimate of the importance of a term in a document. Balanced values of k1 and b parameters were identified, leading to the algorithm's optimal speed and accuracy in assessing word importance considering the input data's peculiarities. The balanced setting of the BM25 algorithm parameters explains the obtained results. They can be used for automated recognition and analysis of news and information on social media based on natural language processing. However, in practice, the effectiveness of the set of parameters depends on linguistic variations, content, and the theme within new input data sets

Financial Reporting Quality During COVID-19 Pandemic: International Evidence

This paper examines the effect of the COVID-19 pandemic on financial reporting quality proxied by earnings conservatism and value relevance of accounting information. Our final sample consists of 10,510 firm-year observations from 29 countries with a balanced number of 5,255 observations for pandemic and pre-pandemic periods. The results show that during the pandemic, the financial statements of companies with a financial year end of 31 March 2020 have different qualities than those in the pre-pandemic period. Lower earnings conservatism is demonstrated in the pandemic period compared to the pre-pandemic period, supporting the argument that business tends to slow the recognition of bad news as the pandemic starts. The value relevance of accounting information declined in the pandemic period, primarily affected by the reduced value relevance of the book value of equity. The findings of this study can help investors evaluate the quality of accounting information and make smarter investment decisions, especially when most companies are in financial trouble.

CFO overconfidence and conditional accounting conservatism

This study investigates the association between Chief Financial Officers (CFOs) overconfidence and conditional accounting conservatism. Relying on upper echelons and overconfidence theories and based on a large sample of US-listed firms’ data from 1992 to 2019 (21,626 firm-year observations), we find a statistically and economically significant negative relationship between CFO overconfidence and conditional accounting conservatism, suggesting that overconfident CFOs tend to diminish conditional accounting conservatism. These findings persist in a series of robustness tests. In the mechanism analysis, we predict that overconfident CFOs aim to convey private information by reducing conditional accounting conservatism. We prove this conjecture by observing that overconfident CFOs who adopt lower levels of conditional accounting conservatism increase earnings informativeness (i.e., the amount of information about future cash flows or earnings contained in current stock returns) and reduce their precautionary incentives to save cash. We further rule out another mechanism (i.e. compensation concerns) that may motivate overconfident CFOs to reduce conditional accounting conservatism. Moreover, we show that overconfident CFOs with higher powers are more able to minimize conditional accounting conservatism. Our study highlights the significance and motivation of overconfident CFOs in determining asymmetric recognition of good and bad news.

Impact of Accounting Conservatism on the Informativeness of Current Stock Prices for Future Earnings

Purpose: The purpose of this paper is to examine how accounting conservatism affects the relation between current stock returns, current earnings, and future earnings.
 Design/methodology/approach: To achieve the purpose of this paper, we divide accounting conservatism into con-ditional and unconditional conservatism and investigate whether the relation between current stock returns, current earnings, and future earnings is different depending on the type of conservatism.
 Findings: We find that the current stock returns of firms with strong conditional conservatism contain more in-formation about future earnings. However, our results show that informativeness of current stock returns on future earnings does not change under unconditional conservatism. 
 Research limitations/implications: Even though we use prior research to capture the proxy for accounting con-servatism, the extent or magnitude of accounting conservatism may not be properly measured by our proxies. However, despite of these limitations, our results still have the following meanings.
 Originality/value: Our results suggest that compared to unconditional conservatism, conditional conservatism re-flects managerial discretion over timely recognition of bad news and results in faster recognition of the impact of future earnings on current earnings. More specifically, our result indicates that conditional conservatism decreases the informativeness of current stock returns for future earnings due to asymmetric timeliness. And our result also provides indirect evidence that conditional conservatism reflects managerial discretion over timely loss recognition compared to unconditional conservatism, and thereby, deteriorates the informativeness of current returns for future earnings.

AAFNDL - An Accurate Fake Information Recognition Model Using Deep Learning for the Vietnamese Language

On the Internet, "fake news" is a common phenomenon that frequently disturbs society because it contains intentionally false information. The issue has been actively researched using supervised learning for automatic fake news detection. Although accuracy is increasing, it is still limited to identifying fake information through channels on social platforms. This study aims to improve the reliability of fake news detection on social networking platforms by examining news from unknown domains. Especially, information on social networks in Vietnam is difficult to detect and prevent because everyone has equal rights to use the Internet for different purposes. These individuals have access to several social media platforms. Any user can post or spread the news through online platforms. These platforms do not attempt to verify users or the content of their locations. As a result, some users try to spread fake news through these platforms to propagate against an individual, a society, an organization, or a political party. In this paper, we proposed analyzing and designing a model for fake news recognition using Deep learning (called AAFNDL). The method to do the work is: 1) First, we analyze the existing techniques such as Bidirectional Encoder Representation from Transformer (BERT); 2) We proceed to build the model for evaluation; and finally, 3) We approach some Modern techniques to apply to the model, such as the Deep Learning technique, classifier technique and so on to classify fake information. Experiments show that our method can improve by up to 8.72% compared to other methods.

The bright side of CEO narcissism and its impact on accounting conservatism

PurposeThis study sheds new light on the personality trait and provides evidence regarding the relation between narcissism and desirable accounting practices, specifically the impact of CEO narcissism on accounting conservatism.Design/methodology/approachThe authors test the relation between CEO narcissism and accounting conservatism for a sample of 907 US companies and their corresponding CEOs for the period between 2010 and 2018. The authors apply three established models of accounting conservatism and measure executives' narcissism using a non-intrusive approach ubiquitous in the literature.FindingsThe authors find that CEO narcissism is associated with speculative accounting practices in the form of timely recognition of positive news and more prudent financial reporting of anticipated negative news. The authors provide the first empirical evidence that, despite its well-known negative effects on corporate financial reporting, executive narcissism can also produce positive outcomes.Originality/valueWhile managerial overconfidence has received much attention, the effects of executives' narcissism are still widely unexplored (Chatterjee and Hambrick, 2007). The authors thus contribute to the literature by investigating the relationship between CEOs' narcissism and accounting conservatism. The authors conjecture CEO narcissism should have a twofold effect on prudent financial reporting. On the one hand, CEOs' narcissism should be associated with low levels of unconditional conservatism due to excessively fast good news recognition. On the other hand, narcissistic executives should be associated with early recognition of negative news and hence with higher levels of conditional conservatism.

A prosocial fake news intervention with durable effects

The present online intervention promoted family-based prosocial values—in terms of helping family members—among young adults to build resistance against fake news. This preregistered randomized controlled trial study is among the first psychological fake news interventions in Eastern Europe, where the free press is weak and state-sponsored misinformation runs riot in mainstream media. In this intervention, participants were endowed with an expert role and requested to write a letter to their digitally less competent relatives explaining six strategies that help fake news recognition. Compared to the active control group there was an immediate effect (d = 0.32) that persisted until the follow-up four weeks later (d = 0.22) on fake news accuracy ratings of the young, advice-giving participants. The intervention also reduced the bullshit receptivity of participants both immediately after the intervention and in the long run. The present work demonstrates the power of using relevant social bonds for motivating behavior change among Eastern European participants. Our prosocial approach with its robust grounding in human psychology might complement prior interventions in the fight against misinformation.

The role of accounting regulation on good news recognition and conditional conservatism: evidence from China’s 2006 accounting regulation

PurposeThis study aims to investigate the impact of accounting regulation (AR) on good news (GNs) recognition and conditional conservatism (CC), using the Chinese experiment in the 2006 AR.Design/methodology/approachThe current study applies the Basu (1997) model to a sample of 26,755 firm-year observations from 1997 to 2017.FindingsThe study finds that GNs recognition is significantly reduced and becomes significantly less than bad news recognition under China’s 2006 AR, suggesting that CC is enhanced under China’s 2006 AR.Research limitations/implicationsThis paper excludes the financial sector from the sample. Thus, generalizing results on the financial sector is not appropriate. In addition, the available literature on this topic is limited as it deals with convergence with IFRS, not the full adoption of IFRS.Practical implicationsThe study findings are expected to benefit policymakers and accounting standard setters in controlling GNs recognition and CC as well.Originality/valueWhereas prior studies on this topic report their results on CC only, without considering the role of AR on GNs recognition, this paper highlights the role of AR on both GNs recognition and CC as well.

An Intelligent System for Detecting Fake News

A significant task facing machine learning and natural language processing is real-time recognition of fake news generated automatically. Social media platforms contribute to the spread of uncertain information, context in which people of different backgrounds all over the world interact. The purpose of this work is to demonstrate the significant role of artificial intelligence in the remarkable generation of the content (here, content with a low degree of trust). As a result of this survey, after identifying and analyzing main research trends in fake news detection, current opportunities are highlighted where specific recommendations could be exploited as solutions for users, especially, on social media. Moreover, the experiments show that fake news can be generated easily through various interventions into the true news. It is about the fact distortion, subject-object exchange and cause confounding. Also, this work highlights the power of generating news with suspicious content using different classifiers of fake like Fakebox and the highly publicized ChatGPT.

Robust Semi-Supervised Fake News Recognition by Effective Augmentations and Ensemble of Diverse Deep Learners

Robust Semi-Supervised Fake News Recognition by Effective Augmentations and Ensemble of Diverse Deep Learners

Meta-heuristic Searched-Ensemble Learning for fake news detection with optimal weighted feature selection approach

Nowadays, fake news has turned into a major problem because of the negative impact that it creates on society. Social media allows people to spread information on the internet with slight investigations and to add fewer filters than the actual content. Nowadays, the false news in the internet community is unsure, and it creates a wrong impression among the users. Detecting false news has become a critical task based on shared content. To tackle the false news growth in social media, various automatic detection schemes were evaluated. The “Natural Language Processing (NLP)” method also gives a prominent solution for false news detection. The main intention of this paper is to design and introduce an innovative false news recognition method using Meta-heuristic Searched-Ensemble Learning (MS-EL). Further, the selected features are extracted by the “Term Frequency-Inverse Document Frequency (TF-IDF)” and also Word2vec features. Here, the extracted selected features are integrated with the Hybrid Squirrel–Dragonfly Search Optimization (HS-DSO) is used to optimize the weighted feature selection approach with the fitness function of solving data variance and correlation. The proposed MS-EL is adopted in the classification part, having three sets of classifiers, Long Short-Term Memory (LSTM), Support Vector Machine (SVM), and Deep Neural Network (DNN). Here, the ensemble classifier is enhanced by the same HS-DSO that shows the parameter tuning with a high convergence rate. From the experimental outcomes, the accuracy of HS-DSO-MS-EL is 22% higher than BMO-MS-EL, 24% higher than SP-BMO-MS-EL, 30% higher than SSA-MS-EL, and 29% higher than DA-MS-EL. Thus, the experimental analysis with standard datasets establishes that the introduced fake news detection method has gained higher accuracy than the existing models.

The influence of feedback on detection and recognition memory of fake news: an experimental study from the perspective of signal detection theory

The objective of this research was to analyze the effect of immediate feedback on the detection and recognition memory of false and true news, using the Signal Detection Theory. For this, two experiments were carried out; in experiment one no feedback was presented; in experiment 2 immediate feedback was included. The experiments were divided into 2 phases: discrimination and news recognition memory. It included the participation of 93 young people, who judged the veracity or falsity of the news using a forced choice task, in 24 news with two types of attributes, physical and non-physical. The results indicated in experiment one, that the non-physical attributes alter the detection and memory recognition of the news, making the physical attribute of the font color irrelevant. In experiment 2, immediate feedback was found to improve detection, but not recognition, which is impaired by non-physical attributes as in experiment one. These findings show the need to present news readers with different sources of immediate verification to improve their detection, as well as the need to generate good information verification practices in the communication sector.

Recurrent Neural Networks with Continuous Learning in Problems of News Streams Multifunctional Processing

The main task of using neural networks is the prompt and accurate solution of various creative tasks, including the analysis and synthesis of news flows, while maintaining the continuity of learning. The result of such processing can be digests, filtered news streams, as well as event forecasts that allow for proactivity in management decisions. Known methods of news processing by neural networks and technical solutions that implement them do not fully provide a solution to the problems that arise in this area. It is necessary to expand their functionality, and improve the space-time signal binding in recurrent neural networks. When processing news flows, simultaneously with continuous training of recurrent neural networks, selection, recognition, restoration, prediction and synthesis of news should be carried out. To reduce the severity of the problem, a promising method of multifunctional processing of news flows is proposed using recurrent neural networks with a logical organization of layers and continuous learning. The method is based on the development of associative processing of textual information in streaming recurrent neural networks with controlled elements. The key features of this method are the multifunctional processing of information flows with changing laws of news appearance. The method provides for operational selection, recognition, restoration, forecasting and synthesis of news based on deep associative continuous processing of links between text elements. The neural network system that implements the proposed method differs from the known solutions by new elements, connections between them, as well as by the functions performed. The results of the experiments confirmed the extended functionality of the method. New features of processing news texts by streaming RNNs are revealed. The proposed solutions can be used to create a new generation of intelligent systems not only for word processing, but also for other types of information.

Experimenting the effect of using visual multimedia intervention to inculcate social media literacy skills to tackle fake news

This study experimented the effect of using visual multimedia intervention to improve users’ social media literacy skills to combat fake news. We carried out a quasi-experiment in one public university in Nigeria and randomly divided 470 participants into equal parts to form a control and experimental group. The respondents in the experimental group were exposed to 8 weeks of training using visual multimedia to improve their social media literacy skills to fight fake news. We realised that those exposed to social media literacy skills training via visual multimedia demonstrated a better knowledge of social media, better recognition of fake news, a higher tendency to verify information and a lesser inclination to share false news. Implications for research and practice were discussed.

News literacy, fake news recognition, and authentication behaviors after exposure to fake news on social media

The global problem of online disinformation has led scholars, educators, and other stakeholders in societies to emphasize the utility of news literacy to engender more critical news audiences. Using a survey among a representative online sample of citizens in Hong Kong ( N = 1485), this study examined how dispositional news literacy was related to individuals’ ability to discern real and fake COVID-related news on social media and their news authentication behaviors. Results showed that higher news literacy was related to greater ability to discern the veracity of real and fake news headlines; greater likelihood of certain internal acts of authentication when exposed to fake news (e.g. assessing content characteristics of the message); and greater likelihood to search online to verify fake news. The findings demonstrated the normative benefits of high dispositional news literacy among the general populace that can attenuate the effects of online disinformation.

A quasi experiment on how the field of librarianship can help in combating fake news

This study experimented how fake news could be combated through social media literacy skills training. A quasi-experiment in one public university's library was conducted and 470 participants were randomly split into equal parts to form a control and experiment group. The respondents in the experiment group were exposed to 8 weeks of training to improve their social media literacy skills to fight fake news. We realised that social media literacy skills training is effective in increasing social media knowledge, users' recognition of fake news, tendency to verify information and reduce the inclination to share false news. Implications for research and practice were discussed.

FAKE NEWS DETECTION USING DEEP LEARNING

Most of the smart phone users prefer to read the news via social media over internet. The news websites are publishing the news, providing the source of authentication. Human inefficiency to distinguish between true and false facts poses fake news as a threat to logical truth, which deteriorates democracy, journalism, and credibility in governmental institutions. In the wake of emerging technologies, there is dire need to develop methodologies, which can minimize the spread of fake messages or rumours that can harm society in any manner. Online clients are normally vulnerable and will, in general, perceive all that they run over Web- based networking media as reliable. Consequently, mechanizing counterfeit news recognition is elementary to keep up hearty online media and informal organization. . It is harmful for the society to believe on the rumours and pretend to be a news. The need of an hmy is to stop the rummys especially in the developing countries, and focus on the correct, authenticated news articles. And so, I propose a model for recognizing forged news, which is a computational stylistic analysis based on natural language processing, efficiently applying deep learning algorithms like ANN algorithm to detect fake news in texts extracted from social media.