News Texts Research Articles

Blending News Text and Economic Policy Uncertainty to Forecast the Company’s Unexpected Earnings

Blending News Text and Economic Policy Uncertainty to Forecast the Company’s Unexpected Earnings

The Effect of ChatGPT-Based Project-Based Learning Model and Digital Literacy on News Text Writing Skills

The Effect of ChatGPT-Based Project-Based Learning Model and Digital Literacy on News Text Writing Skills

U.S. War Correspondents Tweeting Ukraine: A Case Study in Transnational Meta-Journalistic Discourse

ABSTRACT This paper investigates how war correspondents working for U.S.-based news organizations Tweeted about the early stages of the 2022 war in Ukraine, focusing particularly on instances when these war reporters contributed to a distinctly transnational version of what Matt Carlson has termed “metajournalistic discourse” (2016). Defining this concept as the “public expressions evaluating news texts, the practices that produce them, or the conditions of their reception” (Carlson [2016]. “Metajournalistic Discourse and the Meanings of Journalism: Definitional Control, Boundary Work, and Legitimation.” Communication Theory 26 (4): 349–368 , 353), we argue that from February to May of 2022, U.S. war correspondents constructed a discourse that situated their own labor within the boundaries of what counts as the most acceptable form of war journalism, representing their reportage as the most independent and transparent form of war reporting. Conversely, they situated the work of Russian and Ukrainian journalists outside this boundary. The paper ultimately argues that journalism scholars should think more transnationally about the discourses that discuss journalistic labor.

News reporting of public speeches in English on-line media: a constructional perspective

The paper argues that news texts split into stories dwelling on events and reports with their organization influenced by the content of documents or public speeches they render. The two levels of arranging news reports – compositional and semantic – are revealed by the comparison of the BBC’s and CBC’s reports on Ukrainian President Zelenskyy’s address to the American Congress of 22 December 2022. The composition of news reports is three-layered: universal, typical of all news texts with the headline, the lead, the post-lead paragraph and the text body; implicit, with the conventional layout accompanied by descriptive, argumentative and background sections; explicit, with additional headings emphasizing the editorial view of the speaker’s ideas. Semantically, news reports are best structured in terms of morphosyntactic constructions, i.e., prefabricated pairings of form and meaning, which relate the content of news reports to the rendered address. It is found that with respect to the proximity to the original speech the constructions used in the news reports form five basic types: identical, equivalent with those in the speechto underscore the affinity of the speaker’s and readers’ worldviews; synonymous, reflecting the original content through meanings close to the new target audience’s worldview; modifying, changing the dependent elements of the author’s original constructions to adjust their meaning to the target audience’s views; generalizing, fitting the speech content into the public worldview; specifying, clarifying the concepts absent from the readers’ worldview.

Digital media environment in wartime. Russian invasion coverage in Ukrainian professional and amateur news media

RESEARCH OBJECTIVE: The aim of this paper is to compare professional and amateur news media (Telegram) coverage of the Russian invasion of Ukraine and to determine the quality of this coverage. THE RESEARCH PROBLEM AND METHODS: The popularity of Tele­gram has been growing since Russia’s full-scale invasion of Ukraine. To describe the digital media environment in wartime, it is important to observe the activity of both professional and amateur media. Webpages containing news texts were exported with Python programs, and news sources were processed using regular expressions. THE PROCESS OF ARGUMENTATION: At the beginning of the digital platform era there were hopes of a symbiotic relationship between professional media and amateur digital platforms, where the latter would provide more diverse content, thus helping some ignored groups to be heard by the media. As for Telegram, it has a growing number of users and channels, but there are almost no restrictions regarding content. Thus, together with some “alternative” and “ignored” voices, production of harmful content on Telegram is observed around the world: hate speech, extremism, racism, conspiracy theories and propaganda. In Ukraine, Telegram is the most popular source of information during the full-scale Russian invasion and Ukrainians trust it, therefore its contribution to informing people should be analysed. RESEARCH RESULTS: The conducted analysis has demonstrated that popular amateur news media mostly fail to provide clear references to news sources in the war period (in March to October 2022, only every third to fifth post contained a clear reference to a news source). On the other hand, the most popular mass media and amateur mass media sources are mostly the same: media resources connected with the president and the government, verified military sources, etc. CONCLUSIONS, INNOVATIONS AND RECOMMENDATIONS: The scarcity of news sources given in Telegram posts confirms previous observations about the low quality of content on Telegram. However, in some cases Telegram channels use reliable official or professional media news sources. Thus, some additional steps are needed to improve the situation: legislative changes for Telegram channels, ownership transparency, further monitoring of content quality, and media literacy campaigns for channels’ audiences.

Analisis Wacana Kritis Berita Kriminal PADA RAKYAT BENGKULU.COM (Studi Berita Kriminal Oknum Polisi Cabuli Remaja Putri)

News is the latest information or information about something that is happening. News is usually presented through print, broadcast, internet or by word of mouth to third parties and the general public. This research aims to find out and analyze the critical discourse of kriminal news on Rakyat Bengkulu.Com (Study of Kriminal News by Police Officers Cabuli Young Women), this research is qualitative research. The results of this research show a critical discourse analysis of kriminal news on Rakyat Bengkulu.Com (Study of Kriminal News by Police Officers Obscuring Young Women) that in the analysis of the kriminal news text on Rakyat Bengkulu.Com, it can be seen that the text reflects a case of alleged immoral acts involving a police officer and a woman. From this analysis, it was found that the news described the case situation, chronology of events, investigation steps, and the victim's request for accountability. Furthermore, in discursive (mesostructural) analysis, it is shown that the mass media has ideological considerations and journalists' work routines in determining how news is presented. In the context of the Rakyat Bengkulu.Com news, this discursive aspect includes the production, dissemination and consumption of news which influences readers' understanding and reactions to the case. Finally, in socio-cultural (macrostructural) analysis, it was found that this case had situational, institutional and social dimensions that influenced various aspects of the victim's life, the responsibilities of the police institution and broader social issues. This shows the role of the media in shaping opinions and responding to relevant social and legal issues.

Integrating Multimodal Deep Learning for Enhanced News Sentiment Analysis and Market Movement Forecasting

This paper presents a novel multimodal deep learning framework for analyzing news sentiments and forecasting market movements by leveraging natural language processing, deep learning, and auxiliary data sources. Traditional methods often rely solely on textual news data, limiting their predictive power due to the complexity and ambiguity of language. Our approach incorporates additional modalities such as stock prices, social media sentiment, and economic indicators to capture a more comprehensive view of market dynamics. We employ a hybrid deep learning architecture that combines convolutional neural networks (CNNs) for text feature extraction, long short-term memory (LSTM) networks for capturing sequential dependencies, and attention mechanisms to selectively focus on the most relevant features. To address data scarcity, we introduce advanced data augmentation techniques, generating synthetic news headlines based on historical stock price movements and sentiment patterns. The proposed system is evaluated on a comprehensive dataset spanning multiple years, including news headlines, stock prices, social media data, and economic indicators. Our method achieves an accuracy of 77.51%, significantly outperforming traditional methods and demonstrating improved robustness and predictive power. This study highlights the potential of integrating diverse data sources and sophisticated deep learning techniques to enhance news sentiment analysis and market movement forecasting.

EXPLAIN THE CONTENT OF DEVELOPING AUTHENTIC ASSESSMENT INSTRUMENTS IN INDONESIAN LANGUAGE LEARNING VIII SMP NEGERI 2 BINJAI

This study aims to, (1) develop the form of authentic assessment instruments in the learning news text of SMP Negeri 2 Binjai, (2) describe the process of developing authentic assessment instruments in the learning news text of grade VIII SMP Negeri 2 Binjai, (3) describe the feasibility of authentic assessment instruments in the learning news text of grade VIII SMP Negeri 2 Binjai. The type of research used in this research is research and development (R&D). The research instruments used in this study were material expert validation questionnaires, languages, and teacher response tests using the Likert scale. The results of this study proved that authentic assessment instruments in class VIII SMP Negeri 2 news text learning were included in the "excellent" and "suitable for use" categories with material expert scores of 77.5%, linguists 75%, and 95% teacher response test. Therefore, authentic assessment instruments in learning class VIII news texts of SMP Negeri 2 Binjai are worth using.

Pembelajaran Berdiferensiasi Dalam Menulis Teks Berita Di SMPN 7 Kota Jambi

The aim of this research is to determine the application of differentiated learning in learning news texts at SMPN 7 Jambi City. The approach in this research uses a qualitative approach with a case study type approach. Data collection was carried out by means of interviews, observation and documentation in the field. The results of this research show that the implementation of differentiated learning at the content stage, namely presenting learning material in various ways, namely text-based and audiovisual, at the process stage, namely grouping students based on learning readiness and interest in learning, at the product stage, namely giving freedom to students to create news texts according to their interests (video, writing, sticking on cardboard), and in the learning environment, teachers create a comfortable environment for students. The implementation of differentiated learning runs according to the four aspects of differentiated learning contained in the teaching module. Keywords: Independent curriculum, Differentiated learning, News text

Anatomy of sovereign yield behaviour using textual news

While the relationship between the information content of macroeconomic news and the behavior of asset prices has been studied extensively in the finance literature, this study provides a new perspective by examining the impact of textual news on sovereign bond yield spreads in an emerging country. This study used bond market news published in newspapers to develop the sentiment scores using a modified word dictionary to unravel news characteristics. A nonlinear regime-shifting regression model of Markov Regime Shifting (MRS) is used to understand the impact of news on sovereign bond yield spreads. The paper results show that textual news sentiment may explain both steepening and flattening of the yield curve, with monetary and fiscal policy news having the most significant impact on yield spread behaviour. The results hold key implications for policymakers, debt fund managers and other market participants.

A Critical Discourse Analysis of Active and Passive Voice in Three News Texts

This paper applies CDA (critical discourse analysis) to display the hidden relationship between language and power in the discourse of the media; it focuses on the issues which are related to the use of passive or active voice, and tries to investigate which forms of voice are used (passive or active). Data have been collected from three different news texts on the subject of Iraq War Logs. The data have been analysed by using the framework model of Fowler et al. [6] and the findings have been discussed within the framework which is provided by Fairclough, N. [4]. The results show that there is a hidden relationship between language and power in the discourse of the media. Access to the media discourse is limited to specific people only, while other people lack this access. Moreover, the types of voice (passive or active) that are used in different texts determine which sentences are more powerful.

Arabic text classification based on analogical proportions

AbstractText classification is the process of labelling a given set of text documents with predefined classes or categories. Existing Arabic text classifiers are either applying classic Machine Learning algorithms such as k‐NN and SVM or using modern deep learning techniques. The former are assessed using small text collections and their accuracy is still subject to improvement while the latter are efficient in classifying big data collections and show limited effectiveness in classifying small corpora with a large number of categories. This paper proposes a new approach to Arabic text classification to treat small and large data collections while improving the classification rates of existing classifiers. We first demonstrate the ability of analogical proportions (AP) (statements of the form ‘x is to as is to ’), which have recently been shown to be effective in classifying ‘structured’ data, to classify ‘unstructured’ text documents requiring preprocessing. We design an analogical model to express the relationship between text documents and their real categories. Next, based on this principle, we develop two new analogical Arabic text classifiers. These rely on the idea that the category of a new document can be predicted from the categories of three others, in the training set, in case the four documents build together a ‘valid’ analogical proportion on all or on a large number of components extracted from each of them. The two proposed classifiers (denoted AATC1 and AATC2) differ mainly in terms of the keywords extracted for classification. To evaluate the proposed classifiers, we perform an extensive experimental study using five benchmark Arabic text collections with small or large sizes, namely ANT (Arabic News Texts) v2.1 and v1.1, BBC‐Arabic, CNN‐Arabic and AlKhaleej‐2004. We also compare analogical classifiers with both classical ML‐based and Deep Learning‐based classifiers. Results show that AATC2 has the best average accuracy (78.78%) over all other classifiers and the best average precision (0.77) ranked first followed by AATC1 (0.73), NB (0.73) and SVM (0.72) for the ANT corpus v2.1. Besides, AATC1 shows the best average precisions (0.88) and (0.92), respectively for the BBC‐Arabic corpus and AlKhaleej‐2004, and the best average accuracy (85.64%) for CNN‐Arabic over all other classifiers. Results demonstrate the utility of analogical proportions for text classification. In particular, the proposed analogical classifiers are shown to significantly outperform a number of existing Arabic classifiers, and in many cases, compare favourably to the robust SVM classifier.

Digital Literacy Model through Critical Reading of Hoax News Texts with SQ3R Technique

Digital Literacy Model through Critical Reading of Hoax News Texts with SQ3R Technique

The link between translation difficulty and the quality of machine translation: a literature review and empirical investigation

We survey the relevant literature on translation difficulty and automatic evaluation of machine translation (MT) quality and investigate whether source text’s translation difficulty features contain any information about MT quality. We analyse the 2017–2019 Conferences on Machine Translation (WMT) data of machine translation quality of English news text translated to eleven different languages (Chinese, Czech, Estonian, Finnish, Latvian, Lithuanian, German, Gujarati, Kazakh, Russian, and Turkish). We find (weak) negative correlation between the source text’s length, polysemy and structural complexity and the corresponding human evaluated quality of machine translation. This suggests a potentially important but measureable influence of source text’s translation difficulty on MT quality.

Text Classification Model Based on Graph Attention Networks and Adversarial Training

Text information on the internet often has a strong sense of immediacy, constantly reflecting societal dynamics and evolving events. This is especially crucial in the field of news text, where the classification and analysis of these immediate and varied text data become essential. Existing text classification models frequently struggle to effectively represent the semantic and local feature information of texts, limiting their effectiveness. The primary challenge lies in improving the representation of both semantic and local feature information in text classification models, which is critical for capturing the nuanced meanings in rapidly evolving news texts. This paper proposes a deep learning-driven framework designed to enhance the effectiveness of text classification models. The method incorporates noise perturbation during training for adversarial training, thereby enhancing the model’s generalization ability on original samples and increasing its robustness. A graph attention network is employed to extract the contextual semantic information of vocabulary from sequential texts. This information is then combined with extracted sentence feature information to enrich the feature representation of the sequence. An attention mechanism is also introduced to extract more critical feature information from the text, thereby deepening the understanding of textual semantic information. Experimental results demonstrate that this method successfully integrates the boundary and semantic information of vocabulary into the classification task. The approach comprehensively and deeply mines the semantic features of the text, leading to improved classification performance.

Retraction Note: Analysis and evaluation on the quality of news text machine translation based on neural network

Retraction Note: Analysis and evaluation on the quality of news text machine translation based on neural network

Do multisource data matter for NGP prediction? Evidence from the G-LSTM model

Precisely predicting natural gas prices (NGPs) is important because it can provide the necessary decision-making basis for energy scheduling, planning and control. However, NGPs are affected by many factors and exhibit the characteristics of nonlinearity and randomness, which makes accurate predictions challenging. Therefore, in this paper, the information gain of multisource data and the global optimization ability of the gray wolf algorithm are used to build a multifactor-driven NGP hybrid forecasting model to improve the prediction performance. First, the emotional tendency and readability of news text are extracted and calculated by using VADER and textstat tools, respectively. Then the network search index is filtered and integrated by using the correlation coefficient method and the CRITIC method to form alternative variables of multisource data (news and search index). Second, the gray wolf optimization algorithm is used to find and determine the best key parameter group in long short-term memory model. Finally, the spot price of natural gas in Henry Hub from March 1, 2012 to February 28, 2022 is selected as the prediction object, and multi-scenario numerical experiments are carried out to verify the effectiveness of the proposed model. The ablation experiment results show that the information gain brought by multisource data can effectively improve the prediction effect of NGPs. Furthermore, the proposed model has the best prediction performance in different scenarios and can be regarded as a promising prediction tool.

Comparative analysis of translation platforms: Accuracy, fluency, and unambiguity from a technological perspective

This study aims to evaluate and compare the performance of leading translation platforms (Youdao, Google, Baidu, and ChatGPT) for translating news texts. The evaluation focuses on three key dimensions: accuracy, fluency, and unambiguity. Each platform was tested by translating a Chinese news article, and their outputs were analyzed for their handling of specialized terminology, linguistic fluency, and cultural adaptability. The findings reveal significant performance differences, reflecting varying levels of technological maturity, data quality, and cultural adaptability among the platforms. ChatGPT emerged as the top performer due to its high accuracy and fluency, while Youdao, Baidu, and Google exhibited varying strengths and limitations in handling complex news texts. Ultimately, the results highlight the importance of continuous technological innovation and theory-practice integration in improving translation quality, particularly for accurate and effective global news dissemination.

Exploring the Value of Computational Methods for Metajournalistic Discourse: The Example of COVID-19 Reporting in Dutch Newspapers

ABSTRACT The COVID-19 pandemic raised questions about trust in journalism and the quality of news reporting during societal crises. While journalists and media professionals frequently offered critical reflections based on personal experiences and observations, computational methods are not widely used to support these evaluative processes. We aim to extend the conversation on metajournalistic discourse by considering the inclusion of empirical methods for monitoring journalistic practices. By disclosing our findings about Dutch news media’s corona reporting between 2019 and 2022, we demonstrate how computational methods for content analyses of news texts can yield empirically informed insights into different facets of journalistic performance. The corpus includes 106,616 corona-related articles from national and regional newspapers in the Netherlands. We deployed text analytical methods such as topic modelling and named entity recognition to explore Dutch corona reporting in respect to different normative criteria (informing, monitoring, offering platforms for discussion and opinion, interpretation, analysis, and setting public agendas). The study was requested by a large Dutch newspaper to receive a systematic-empirical analysis of journalistic practice for self-evaluation. We argue that computational methods combined with qualitative analyses can stimulate dialogue and critical reflection among news media professionals.

An Ensemble Keyword Extraction Model for News Texts with Statistical and Graphical Features

Keyword extraction is an essential tool for many text mining applications such as automatic indexing, summarizing, classification, clustering and automatic filtering. Automated keyword extraction is essential as the daily text data to be reached and processed have increased tremendously over the Internet, e.g. millions of news articles are published daily online. In this paper, a novel ensemble model for automatic extraction of keywords from news articles is proposed. The proposed model handles keyword extraction as a sequence labeling task. Two sub-modules representing the statistical and graphical features by their calculated scores for each input token were combined in the token classification module. The Ensemble Token Classification module was trained and tested separately with the ensemble algorithms Random Forest, XgBoost, Decision Tree and Voting Classification. For training, we collected two news datasets from Kazakh and Russian news sites published in Cyrillic alphabet. We also collected an Arabic news dataset, ArabianNews. The performance of the model was also compared with the widely used 500N-KPCrowd dataset in the literature, which consists of English news content in Latin alphabet. The proposed model achieved the best performance with an [Formula: see text]-score of 0.71 and 0.86 on the 500N-KPCrowd and Russian datasets, respectively. We attained the best [Formula: see text]-score (0.97) with the KazakhNews and ArabianNews datasets.