News Texts Research Articles

Does Peer‐to‐Peer Lending Have Resilience During the COVID‐19 Pandemic? Evidence From China

ABSTRACTDoes Peer‐to‐Peer (P2P) lending, a digital FinTech platform, show resilience in response to COVID‐19 shocks? We examine this question by considering both the absorbing and adapting abilities based on psychological perceptions and objective infection cases. We collect data on infection cases, the Baidu search index, and 54,166 textual news articles to propose a novel perspective that incorporates both subjective psychological perceptions and objective infection cases to measure the response of P2P lending firms to COVID‐19 shocks. We calculate the extent of P2P lending's resilience using four metrics: maximum loss, total loss, recovery periods, and net impacts from local projections impulse responses. Our findings demonstrate that P2P lending displays resilience, as its total trade volume, new investors, and new investments initially decrease slightly and then rebound in response to the COVID‐19 shock. Psychological perception of COVID‐19 shocks, rather than objective infection, elicits a more pronounced response in the performance of P2P lending firms.

Classification of human- and AI-generated texts for different languages and domains

AbstractChatbots based on large language models (LLMs) like ChatGPT are available to the wide public. These tools can for instance be used by students to generate essays or whole theses from scratch or by rephrasing an existing text. But how does for instance a teacher know whether a text is written by a student or an AI? In this paper, we investigate perplexity, semantic, list lookup, document, error-based, readability, AI feedback and text vector features to classify human-generated and AI-generated texts from the educational domain as well as news articles. We analyze two scenarios: (1) The detection of text generated by AI from scratch, and (2) the detection of text rephrased by AI. Since we assumed that classification is more difficult when the AI has been prompted to create or rephrase the text in a way that a human would not recognize that it was generated or rephrased by an AI, we also investigate this advanced prompting scenario. To train, fine-tune and test the classifiers, we created the Multilingual Human-AI-Generated Text Corpus which contains human-generated, AI-generated and AI-rephrased texts from the educational domain in English, French, German, and Spanish and English texts from the news domain. We demonstrate that the same features can be used for the detection of AI-generated and AI-rephrased texts from the educational domain in all languages and the detection of AI-generated and AI-rephrased news texts. Our best systems significantly outperform GPTZero and ZeroGPT—state-of-the-art systems for the detection of AI-generated text. Our best text rephrasing detection system even outperforms GPTZero by 181.3% relative in F1-score.

Research on the Application of Improved BERT‐DPCNN Model in Chinese News Text Classification

ABSTRACTThis paper introduces an enhanced BERT‐DPCNN model for the task of Chinese news text classification. The model addresses the common challenge of balancing accuracy and computational efficiency in existing models, especially when dealing with large‐scale, high‐dimensional text data. To tackle this issue, the paper proposes an improved BERT‐DPCNN model that integrates BERT's pre‐trained language model with DPCNN's efficient convolutional structure to capture deep semantic information and key features from the text. Additionally, the paper incorporates the zebra optimization algorithm (ZOA) to dynamically optimize the model's hyperparameters, overcoming the limitations of manual tuning in traditional models. By automatically optimizing hyperparameters such as batch size, learning rate, and the number of filters through ZOA, the model's classification performance is significantly enhanced. Experimental results demonstrate that the improved ZOA‐BERT‐DPCNN model outperforms traditional methods on the THUCNEWS Chinese news dataset, not only verifying its effectiveness in news text classification tasks but also showcasing its potential to enhance classification performance.

СХЕМОСТРАТЕГІЇ БУДОВИ АНГЛОМОВНИХ ТЕКСТІВ НОВИН: ЛІНГВОКОГНІТИВНИЙ РАКУРС

The paper specifi es the notion of communicative strategies widespread in modern linguistics suggesting the concept of schematic strategies treated as a means of semiotic modifi cation of the addressee’s conceptual structures making up the worldview common for the speakers of a particular language. Drawing on the BBC’s English news texts about the Russia-Ukraine war, the article singles out three groups of moves meant to implement schematic strategies. Referential moves of news texts activate the mental structures representing individuals, social groups, and states or their confrontation. Status moves distribute the textual content with respect to its topicality in the headline, the lead, the sections of current and previous events, and of comments and context. Accentuating moves as an implicit means of infl uencing the audience through reference to objects in the most prominent news text positions of paragraph beginnings and endings form particular subtypes. The move of initial focusing foregrounds the units denoting the same referent in the headline and the subsequent p aragraphs emphasizing its importance. The move of fi nal focusing concentrates the reader’s attention on the referents denoted at the ending of the headline and of the following paragraphs in the texts on politicians’ speeches while the units in the paragraphs’ initial positions refer to the key ideas of the talks under discussion. The move of fi nal-initial focusing foregrounds in the subsequent paragraphs of a news text the units denoting a referent at the end of the headline contrasting it with the one foregrounded in its beginning. The defocusing move names referents in the background positions of a news text bringing down their importance while giving prominence to the focal elements. The identifying move indicates in the lead the referents missing from the headline with the organization of the rest of the text drawing on the moves outlined above. The multidirectional move foregrounds in the headline and each subsequent paragraph different units denoting referents comprising the mental structure of confrontation: its source, target, and the results of their counterforce. Key words: schematic strategy, communicative strategy, move, cognitive linguistics, English news text, Russia-Ukraine war, influence.

Identifying issues in international construction projects from news text using pre-trained models and clustering

The uncontrollable nature of international construction projects requires continuous monitoring of issues in the host country. News articles can provide relevant information to monitor the issues, but the manual investigation of substantial news text is impractical. This paper proposes a framework to automatically collect information related to the host country's business environments from news text, consisting of three modules: (1) web scraping for collecting news text; (2) text embedding using a pre-trained language model; and (3) text clustering for extracting essential issues. Applying this framework to real-world news demonstrated its proficiency in identifying significant issues, outperforming the existing similar methods in terms of: (1) the accuracy of issue identification; (2) the quality of identified issues; and (3) the degree of human intervention. This paper contributes to the body of knowledge by showcasing the utility of news text in gathering information and identifying issues about the host country during international projects.

Framing Pro-Palestine Protests: A Comparative Analysis of News Coverage by The Guardian and USA Today

This study examines the framing strategies employed by The Guardian and USA Today in their coverage of the pro-Palestine protests in American universities during April and May 2024. Accordingly, the corpus comprises 213 news texts, with The Guardian contributing 106 articles and USA Today contributing 107 articles from April 17 to May 17, 2024. The research identifies and contrasts the dominant frames used by these two newspapers by conducting a framing content analysis on the selected news articles. The Guardian frequently utilizes conflict and adversarial frames, emphasizing university-wide ideological and political struggles. Conversely, USA Today focuses on broader political confrontations and administrative accountability, often highlighting institutional responses and governance issues. The analysis process is guided by framing theory to reveal how news representations influence public perception and interpretation and media agenda-setting theory to explore how media news coverage prioritizes specific issues within the same event. The findings underscore the decisive role of media framing in shaping public opinion and priorities, particularly in the context of global and political issues.

What Comes After the Algorithm? An Investigation of Journalists’ Post-editing of Automated News Text

ABSTRACT Journalists often perceive that automated journalism produces news texts that lack narrative and editorial quality. Therefore, they sometimes manually edit automated output before publication, creating so-called “post-edited” variants. Using an exploratory sequential mixed-methods design, this study aims to advance research on automated journalism by investigating the steps journalists say they take when post-editing and testing whether—and if so, how—these steps are reflected systematically on a larger scale in post-edited news texts. First, journalists’ statements about the post-editing process were gathered in semi-structured interviews. Then, qualitative content analysis of automated news stories and their post-edited offspring examined whether post-edited stories contain evidence of forms of editing not mentioned by the journalists. Second, the qualitative findings were investigated quantitatively using comparative content analysis (N = 282). The qualitative findings suggest that a range of editorial steps may be taken during post-editing. The quantitative analysis shows that some of these steps are reflected systematically in post-editing, resulting in significant differences between automated news stories and their post-edited offspring. However, findings also show differences between journalists’ reports and post-editing practice on a larger scale, including evidence that forms of editing were taking place that were not mentioned in the interviews.

Text Spillover: Measuring connectedness of financial institutions based on news text data

The interconnectedness of financial institutions can facilitate the contagion of financial crises and poses significant risks to financial stability. While well-established econometric spillover measures exist, they are usually based on numerical data, which may not always be available or timely. This paper explores the feasibility of using text data analysis as an alternative measure of interconnectedness between financial institutions. A novel news-based spillover measure is constructed and applied to a system of five banks, using an up-to-date dataset of nearly 47,000 news articles. The results of this text spillover index are compared to the Diebold-Yilmaz spillover index for validation. The findings indicate that text data analysis can provide valuable insight into the interconnectedness of financial institutions and substantiate the underlying hypothesis that the co-occurrence of actors in news articles allows direct conclusions about their connectedness. This enables new types of analysis and extends spillover measures to a wider range of actors

Analysis of Cohesion and Coherence of Middle Eastern News Discourse and its Implications for Indonesian Language Learning in High School

Cohesion and coherence are parts that form a text/discourse. Both of them characterize the completeness of a text/discourse. This research aims to describe 1) the form of cohesion of Middle Eastern news discourse in the June 2024 edition of Sindonews.com and its implications for Indonesian language learning in high school, 2) the form of coherence of Middle Eastern news discourse in the June 2024 edition of Sindonews.com and its implications for Indonesian language learning in high school, and 3) the implications of the results of the analysis of the form of cohesion and coherence of Middle Eastern news discourse in the June 2024 edition of Sindonews.com and its implications for Indonesian language learning in high school. This research is qualitative descriptive research. The subject of this research is news text about the Middle East in the June 2024 edition of Sindonews.com. The results of this research are 1) the form of cohesion contained in the news discourse about the Middle East in the June 2024 edition of Sindonews.com in the form of lexical and grammatical forms. In grammatical cohesion, the dominant parts are references and conjunctions. Meanwhile, the dominant part of lexical cohesion is repetition. 2) the form of coherence contained in the news discourse about the Middle East in the June 2024 edition of Sindonews.com, namely cause and effect, conditions for results, applicative, identification, specific generic, background to conclusions, and means of results. Meanwhile, the dominant coherence relationship is the cause-and-effect relationship. 3) The implication for learning is that this form of cohesion and coherence in news texts can be used as teaching material for Indonesian language subjects, news text material in high school.

Aftermath of the Fourth Taiwan Strait Crisis: Media Discourse Analysis of Taiwan and India’s Perception of China’s Threat

This study examines Taiwan and India’s media discourse and political elite statements from 1 July 2022 to 1 October 2023, starting from the fourth Taiwan Strait crisis in 2022. We conducted a discourse analysis of over 121,901 news texts using extensive data analysis and text-mining software. Additionally, our research team conducted expert focus group discussions with renowned think tanks and research centres in New Delhi, India. Building upon the data text analysis findings, we further explored India’s stance on the Taiwan Strait conflict, its perception of the Chinese threat and the impact of the ongoing ‘new normal’ crisis on India’s Indo-Pacific strategic layout. The aim was to uncover common security needs and intersecting interests between the two countries, considering their different national interests and geopolitical considerations. Based on the analysis of textual mining and expert discussions, this study reveals the following: First, India is concerned about the current situation in the Taiwan Strait. Second, the current tensions in the Taiwan Strait will influence India’s strategic layout. Third, there is a growing consensus within India regarding the Chinese threat. In light of these findings, this study suggests that Taiwan and India share interests in democratic values, maritime traffic security and high-tech industry supply chains. Taiwan possesses expertise in China studies and a deep understanding of China’s military threat, while Taiwan and India possess specific technological capabilities. Based on these foundations, there are potential opportunities for further dialogue, exchange and cooperation in relevant fields between India and Taiwan.

The effect of using AI and Google Translate on translating BBC media texts into Arabic

Recent post-human translation studies increasingly explore the intricate relationship between artificial intelligence (AI) and translation. This paper examines the complexities of AI, with a focus on machine translation (MT), and its transformative effect on translation in light of technological advancements. Such transformation has led to new roles and skills for translators. Our research primarily scrutinizes the style and messaging quality of English-Arabic media text translations. We aim to illuminate how AI and MT challenge the translation industry, discussing both potential benefits and risks for language professionals, and presenting a theoretical framework for the changing roles of human translators. This study specifically identifies common errors in MT for Arabic-English news texts and vice versa, assessing translation quality, fluency, semantic accuracy, and the need for human intervention in corrections. We selected purposeful examples from online newspapers and used a qualitative method for data analysis, identifying errors based on Hsu's (2014) machine translation error classification.

Sensing Facts and Emotions: Intelligent Technology's Redefinition of News Text Presentation

Intelligent technology represented by artificial intelligence has profoundly changed the boundaries and patterns of journalism. Traditional media reports are constrained by conditions such as the shape of the newspaper layout, presenting independent and objective content of individual reports. Intelligent technology has changed this pattern, the presentation of news text will reflect the characteristics of "information layer of basic facts" and "opinion layer of emotional analysis" co-presentation, while adhering to the separation of the facts and opinions of the report itself, the typical attitudes and emotions of the user feedback and the facts of the media report are jointly presented to the public. While maintaining the separation of facts and opinions in the report itself, the typical attitudes and emotions of user feedback are presented to the public together with the facts of the media report. This will present the public with an intuitive and visualized opinion environment through sentiment and text analysis while quickly distilling and summarizing the main facts in the information flood. At the same time, intelligent technology further shortens the time lag between the occurrence of an event and the release of a news report, shortening the time lag and the perception of the importance of time lag in a cyclical manner, shaping the practitioner's view of time lag.

Instructional Design of Problem-Based Learning Models: A Strategy for Improving Students' Critical Thinking Abilities in Learning News Texts

Instructional Design of Problem-Based Learning Models: A Strategy for Improving Students' Critical Thinking Abilities in Learning News Texts

Pembingkaian Media Dalam Perpektif Hak Asasi Manusia Terhadap Kasus Intoleransi Beragama dan Berkeyakinan di Masa Pandemi

Coverage of the issue of violations Religion and Belief Freedom during the pandemic is not free from the bias of journalists and mass media. The issue, which is reported inaccurately and sometimes tends to sharpen existing differences in society. It should then be able to provide an overview to the public on how to understand differences amidst the pandemic. Detik.com, as a popular mainstream online media in Indonesia, has the advantage of distributing news quickly. Coverage of the issue of violations is quite common in detik.com. This study focuses on examining how the news texts regarding the issue during pandemic are constructed by detik.com using the framing analysis method of Zhongdang Pan and Gerald M. Kosicki. In this study, the researcher uses the theory of the construction of social reality and views it from the perspective of Human Rights. This aims to see how detik.com constructs its news texts regarding the issue during the pandemic. Based on the research results, detik.com tends to be more biased towards the victims, as seen from most prominent structural units emphasized by detik.com when constructing news texts regarding the issue violations, such as headlines, leads, background information, quotes from sources, statements, details, and lexicon.

"News has various shades": Quantifying and Analyzing Media Bias at Aspect-level Granularity

This research presents a novel approach to dissecting media narratives, with a specific focus on aspect-level granularity and variance in addressing multifaceted news topics. Unlike previous studies, which often concentrate solely on ideological or political biases, our methodology delves deeper, exploring how diverse media outlets navigate the complexities of various news aspects. Through a detailed case study on Indian government policies, we uncover distinctive biases and variations in reporting between Indian and international media. Crucially, our methodology leverages Natural Language Inference (NLI) to identify news aspects and ascertain aspect-level sentiment from news text, enabling scalable and precise quantification of bias across diverse media narratives. Our findings illuminate the multifaceted layers of media coverage, revealing nuanced stances on different aspects of the same topic and the dynamic nature of biases over time. Importantly, our comprehensive framework quantifies media bias through data-driven quantitative evaluation, capturing both selection bias based on news aspect coverage and statement bias based on sentiment polarity. It is relevant to note that in our context, bias is not inherently favourable or unfavourable but rather serves as a quantitative metric to measure divergence in news presentation from the overall average. Thus, our contributions not only enhance understanding of media bias but also introduce a methodology distinguished for its comprehensive and scalable approach to analyzing media narratives, with substantial implications for future research and discussions in this domain.

The news values of fake news

Fake news poses a significant threat to society by undermining public trust and consensus on critical issues. Although there is a considerable amount of research on the linguistic features of fake news texts, a comprehensive understanding of how language is used to persuade and promote specific ideologies within them is still lacking. This study addresses this gap by analyzing fake news discourse through the lens of news values. We apply the Discursive News Values Analysis (DNVA) framework and key semantic domain analysis to a corpus of fake news stories on vaccination, climate change, and COVID-19. We identify a set of news values that differentiate fake from mainstream news discourse. Our findings reveal that fake news emphasizes negativity, unexpectedness, consonance, and facticity, while also relying on the previously undocumented news values of subversiveness, causality, religiosity, and historicity. These values form a powerful discursive toolkit exploited by fake news writers to craft compelling false narratives.

Speech Acts of Advising in Selected News Texts about Coronavirus

After the invading of Coronavirus in 2019 in China, The world Health Organization has declared this virus as a pandemic. The virus outbreaks among countries terrifying manner which makes governments take some quick decisions to stop it. Government all over the world employes TV channels and media to deliver instructions and advices to their peoples. But, this time the advice and instructions are different in style and language, since there is a crisis which may destroy the whole country, as what has happened in Ital pandemicly. Therefore, the speech act of advising which is used under this circumstances is a bit different for imposing more emphasis to face that danger. Advising people needs to be more directive and imperative. This study hypothesizes that under serious crisis, speech act of advice tends to be more directive and imperative to make the language receiver takes the advice in his/her consideration. This study tries to answer the following question: Do strategies in performing speech act of advice under a common crisis differ between government and Tv Channels? The study analyses ten texts from two different news channels. The first five texts are taken from Kurdistan 24 Channel and the other five are taken from BBC NEWS Channel. The study reaches to the findings that under serious crisis, there is no difference between western and esteran government in the strategies used to direct their people. In both channels, the speech act of advice tends to be direct and imperative more than other strategies to express it.

Turkish Text Classification Based On Wrapper Feature Selection Using Particle Swarm Optimization

The vast majority of the digital era data is stored as text. Text mining is an integral part of data mining. Text classification (TC) is a natural language processing (NLP) operation often needed in text mining. This operation is needed in numerous kinds of research such as information retrieval, document classification, language detection, sentiment analysis, etc. According to the literature, the filter feature selection methods have often been applied to reduce the dimensionality of data in Turkish TC. However, the wrapper-based feature selection methods can provide better classification accuracies than the filter methods. Motivated by this idea, a Turkish TC method based on wrapper feature selection using particle swarm optimization algorithm (PSO) and multinomial naive bayes (MNB) classifier is proposed in this study. TTC-3600 Turkish news texts are used for TC in the experiments. The proposed method achieves a classification accuracy of 94.55% on TTC-3600 Turkish news text dataset by using stemming Tf-Idf features. Hence, it produces competitive accuracies to the cutting-edge Turkish TC methods.

Indonesian Fake News Classification Using Transfer Learning in CNN and LSTM

Fake news spreads quickly and is challenging to stop due to the ease of accessing and sharing information online. Deep learning techniques are a method that can be used to identify fake news quickly and accurately. The types of neural networks commonly utilized in deep learning architectures include Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM), which can perform well when managing the task of classifying fake news, according to several pertinent studies. Regarding handling instances of Indonesian fake news classification, this study compares how well the CNN and LSTM models perform. However, given that Indonesian is a low-resource language with scant documentation, it is challenging to build an adequate data set. At the same time, the CNN and LSTM classification models require significant training data. We proposed a transfer learning method by combining two classification models with a pre-trained IndoBERT language model. 1340 news text data were used, including 643 actual news texts from CNN Indonesia, Liputan6, and Detik and 697 fake news texts from TurnBackHoax. As a result, the performance of the combination of the LSTM classification model with IndoBERT outperformed that of the CNN classification model with IndoBERT, which only produced an accuracy of 92.91%, down by 6%, and was able to produce an accuracy of up to 97.76%, an increase of 4.8% from before. Furthermore, the results show that the LSTM classification model outperforms the CNN classification model in capturing the representation created by IndoBERT. Additionally, these insights may serve as a basis for future research on identifying fake news in Indonesia, helping to improve methods for combatting misinformation in Indonesia.

Rekonstruksi Kalimat pada Teks Berita di Laman CNN Indonesia Edisi Desember 2023 sebagai Bahan Ajar Membaca Kritis Siswa Kelas 12 SMA

News is a text that has an important role in disseminating information. News text is so useful that it has become a literacy material for students due to the development of advanced technology. Although it has become one of the sources of student literacy, online news texts are still classified as having syntactic errors such as a lack of understanding of the language used in compiling a sentence in the news. The purpose of this study is to describe sentence errors in the news contained in the old CNN Indonesia news, such as "December Sky Phenomena: Geminid Meteor Shower to Solstice", "Mandalika Circuit's Important Role in Bringing Indonesian Drivers to Success in Asia", "Solar Storms Raining on Earth, RI Safe?", "The First Candidate Debate will be Held at the KPU Office on December 12", and "When HIV Tests Become a Scary Scourge Because of People's Scorn". This research uses a descriptive method with a qualitative approach. The results of this study found several findings such as effective sentences, ineffective sentences, spelling errors according to EYD, errors in the use of diction, and sentences that are not clear to the reader. The most common error found in the five news articles is ineffective sentences. The benefit of this research is to provide knowledge insights in the form of understanding sentence writing systematically, which will be useful in improving understanding of authorship in news texts and preventing errors from occurring again in the same news so that it is worthy of being a literacy material for 12th grade high school students.