Text Analyses Research Articles

Regulation-induced digitalization

This paper investigates how environmental regulation induces firm digitalization. We construct a digital index based on textual analyses and find that after the implementation of the program, pilot firms' digitalization increased relative to that of a group of carefully matched control firms, which is opposite to the findings in the extant literature on technology adoption. This increase cannot be fully explained by regional unobservables, firms' own innovation, firm selection, or other policies. The results are robust when we consider firm subsidiaries. The increase in digitalization is not due to regulatory arbitrage, and the industry-level concentration of digitalization changes little.

EmotionCast: An Emotion-Driven Intelligent Broadcasting System for Dynamic Camera Switching.

Traditional broadcasting methods often result in fatigue and decision-making errors when dealing with complex and diverse live content. Current research on intelligent broadcasting primarily relies on preset rules and model-based decisions, which have limited capabilities for understanding emotional dynamics. To address these issues, this study proposed and developed an emotion-driven intelligent broadcasting system, EmotionCast, to enhance the efficiency of camera switching during live broadcasts through decisions based on multimodal emotion recognition technology. Initially, the system employs sensing technologies to collect real-time video and audio data from multiple cameras, utilizing deep learning algorithms to analyze facial expressions and vocal tone cues for emotion detection. Subsequently, the visual, audio, and textual analyses were integrated to generate an emotional score for each camera. Finally, the score for each camera shot at the current time point was calculated by combining the current emotion score with the optimal scores from the preceding time window. This approach ensured optimal camera switching, thereby enabling swift responses to emotional changes. EmotionCast can be applied in various sensing environments such as sports events, concerts, and large-scale performances. The experimental results demonstrate that EmotionCast excels in switching accuracy, emotional resonance, and audience satisfaction, significantly enhancing emotional engagement compared to traditional broadcasting methods.

Social media and the negotiation of travel photo authenticity in the making of Wanghongjingdian in China

Social media have reconfigured place-making in tourism so that industry-led place promotion through professionally produced media representations is nowadays intertwined with bottom-up processes involving online influencers and ordinary people. The emergence of wanghongjingdian (i.e. internet-popular tourist attractions) in China is a case in point. This article aims at contributing to our understanding of the production of wanghongjingdian through a focus on the bottom-up negotiation between influential bloggers and ordinary users, especially on the “authenticity” of travel photos. We conducted a virtual ethnography and visual and textual analyses of 290 blogs on Xiaohongshu, a highly popular visual-oriented platform in China. Three authentication strategies, which relate to existing theoretical conceptualizations of authenticity in nuanced manners, are identified, and the negotiation process between influencers and ordinary users is marked by a critical participatory culture, multiplicity of practices, and individuality-in-community. These features help us understand the production of wanghongjingdian on Chinese social media.

Racializing COVID-19: Race-Related and Racist Language on Facebook, Pandemic Othering, and Concern About COVID-19

ABSTRACT To what extent have social media discussions of health racialized during the COVID-19 pandemic? How has the spread of racialized social media health content influenced attitudes about race and COVID-19? To answer these questions, we collected data on public US Facebook posts from 2019–2020 related to health and health care (n ≃ 9,659,000) and document a rise in race-related and racist language in online discussions directly-related to health during the pandemic. We show a substantive increase in Facebook posts connecting health to race and racism beginning around the announcement of the first declared case of COVID-19 in March 2020. Text analyses reveal that posts with explicitly racist language surrounding blame for COVID-19, discussions of unemployment rates and benefits, as well as immigration were shared more on average than the typical post about health generally. We connect the spread of this language on Facebook to survey responses (n ≃ 231,000) from Nationscape. We show that weekly increases in racist and race-related health language on Facebook are related to declines in concern for COVID-19 and heightened anti-Asian attitudes, especially for racially resentful white Facebook users.

GPT is an effective tool for multilingual psychological text analysis

The social and behavioral sciences have been increasingly using automated text analysis to measure psychological constructs in text. We explore whether GPT, the large-language model (LLM) underlying the AI chatbot ChatGPT, can be used as a tool for automated psychological text analysis in several languages. Across 15 datasets (n = 47,925 manually annotated tweets and news headlines), we tested whether different versions of GPT (3.5 Turbo, 4, and 4 Turbo) can accurately detect psychological constructs (sentiment, discrete emotions, offensiveness, and moral foundations) across 12 languages. We found that GPT (r = 0.59 to 0.77) performed much better than English-language dictionary analysis (r = 0.20 to 0.30) at detecting psychological constructs as judged by manual annotators. GPT performed nearly as well as, and sometimes better than, several top-performing fine-tuned machine learning models. Moreover, GPT's performance improved across successive versions of the model, particularly for lesser-spoken languages, and became less expensive. Overall, GPT may be superior to many existing methods of automated text analysis, since it achieves relatively high accuracy across many languages, requires no training data, and is easy to use with simple prompts (e.g., "is this text negative?") and little coding experience. We provide sample code and a video tutorial for analyzing text with the GPT application programming interface. We argue that GPT and other LLMs help democratize automated text analysis by making advanced natural language processing capabilities more accessible, and may help facilitate more cross-linguistic research with understudied languages.

Navigating the new frontier: the impact of artificial intelligence on students’ entrepreneurial competencies

PurposeArtificial intelligence (AI) technologies have led to significant transformations across industries and society, including the field of education. The integration of AI in educational settings has the potential to improve students' learning experience and support their individual competencies when paired with non-AI methods. Despite the growing importance of AI in modern education, there remains a noticeable research gap regarding its use in entrepreneurship education and the effects of Chatbots on students' entrepreneurial competencies. To address this gap, an exploratory study was conducted on undergraduate students who were tasked with using ChatGPT to improve their business model canvas.Design/methodology/approachThe chosen methodology aligned with the research purpose, aiming to explore the relationship between Generative AI and competencies. Due to the novel nature of the research problem, an exploratory study was conducted using a mixed methods approach. A survey with open- and closed-ended questions was designed, and statistical and text analyses were performed to interpret data and test identified propositions.FindingsThe findings of this study indicate that ChatGPT can enhance the types of students' entrepreneurial competencies considered in this study: spotting opportunities, creativity, vision, valuing ideas and ethical and sustainable thinking. The results show that ChatGPT can be particularly helpful to improve the ability of students of valuing ideas.Originality/valueOverall, this study highlights the potential of adopting ChatGPT in experiential learning methodologies for enhancing students' entrepreneurial competencies and improving their learning outcomes.

Mapping the moral foundations of the European Union: Why a lack of moral diversity may undermine perceived EU legitimacy.

The stability and effectiveness of supranational organizations, like the European Union (EU), will improve when citizens perceive them as legitimate. Across three studies, two of which were preregistered, and a preregistered pilot study, we combined EU legal expertise with social psychological theory on morality to understand how perceived EU legitimacy is influenced by a sense of moral alignment with the EU. We propose that, currently, the EU gives more weight to values linked to "individualizing" moral foundations (e.g. compassion, social justice, and equality) than to values linked to "binding" moral foundations (e.g. patriotism, religion, and traditionalism). As this may leave people who endorse binding moral foundations feel unrepresented, we investigated whether the EU could gain legitimacy by appealing to values that resonate with binding moral foundations. In study 1, text analyses revealed that the European Commission President's State of the Union speeches indeed appeal more to individualizing than to binding moral foundations. Study 2 (n = 595) provided correlational evidence that the negative relationship between binding moral foundations and perceived EU legitimacy was mediated by lower moral alignment with EU law. Finally, study 3 (n = 567) showed through an experiment that reframing or rebalancing EU law to better align it with binding moral foundations could increase perceived EU legitimacy among people who endorse these moral foundations. The results illustrate the importance of understanding and attending to moral diversity among EU citizens. More generally, our work shows how a collaboration between social psychology and law contributes to safeguarding the legitimacy of supranational organizations.

Climate disasters and the difficult human/nature relationship

Environmental disasters are concrete examples of the difficult coexistence between man and nature and expose the fragility of ecosystems and the urgent need to rethink the development model of contemporary society. With a socio-political approach, the text analyses the relationship between man and nature and the environmental tragedies caused by global warming, denouncing the development process of industrialised societies and questioning the environment destroyed by the diversity of human actions. It discusses extreme weather events that have resulted in major environmental disasters, both in Brazil and in other countries, and provides quantitative data on the victims of these disasters. It reveals the unequal distribution of income and space that exposes the most vulnerable populations to the rigours of climate change and argues that man's predatory relationship with nature is the driving force behind environmental degradation and is based on the unethical interests of the market. He concludes on the importance of urban planning, the projection of more humane spaces and obedience to environmental protection laws.

Adaptive Evolutionary Computing Ensemble Learning Model for Sentiment Analysis

Standard machine learning and deep learning architectures have been widely used in the field of sentiment analysis, but their performance is unsatisfactory if the input texts are short (e.g., social media posts). Specifically, the accuracy of standard machine learning methods crucially depends on the richness and completeness of the features used to represent the texts, and in the case of short messages, it is often difficult to obtain high-quality features. Conversely, methods based on deep learning can achieve better expressiveness, but these methods are computationally demanding and often suffer from over-fitting. This paper proposes a new adaptive evolutionary computational integrated learning model (AdaECELM) to overcome the problems encountered by traditional machine learning and deep learning models in sentiment analysis for short texts. AdaECELM consists of three phases: feature selection, sub classifier training, and global integration learning. First, a grid search is used for feature extraction and selection of term frequency-inverse document frequency (TF-IDF). Second, cuckoo search (CS) is introduced to optimize the combined hyperparameters in the sub-classifier support vector machine (SVM). Finally, the training set is divided into different feature subsets for sub-classifier training, and then the trained sub-classifiers are integrated and learned using the AdaBoost integrated soft voting method. Extensive experiments were conducted on six real polar sentiment analysis data sets. The results show that the AdaECELM model outperforms the traditional ML comparison methods according to evaluation metrics such as accuracy, precision, recall, and F1-score in all cases, and we report an improvement in accuracy exceeding 4.5%, the second-best competitor.

Exploring climate change discourses on the internet: a topic modeling study across ten years

This study examines online discourses about climate change using automated text analysis. Samples of web contributions from various sources, such as blogs, online newspaper articles, readers’ comments, tweets, and others, were collected in 2013, 2019, and 2023. The major themes people talk about when discussing climate change on the internet were identified using structural topic modeling. The study also explored how people discuss climate change differently when using different types of media channels and how the main themes change over time. Six major topics were identified for each point in time; associations between topics and media types were examined. Three topics emerged repeatedly: Dispute over the reality and the implications of climate change, discussions about aspects of the science of climate change, and discussions about causes and consequences of global warming such as sea level rise. In 2019 and 2023, the topic of action (what to do) and the topic of looming disaster showed up repeatedly. Topics limited to one specific corpus were policy discussions, debates about emissions, about renewable energies, about agricultural implications, and expressions of despair. The topical structure could be organized in two general groups, one group containing more science-related topics, and another group containing expressions of dispute, disagreement, and affect. A significant correspondence between media type and prevalent topic could be detected. Media such as Twitter/X and comments are dominated by dispute and despair expressions, whereas media such as blogs and journalistic online articles are primarily concerned with scientific topics in general, as well as with specific topics concerning global warming or renewable energies. In sum, we found noteworthy topical invariances across ten years, and systematic correspondences between topics and media types. The study contributes to the understanding of online discourses about climate change and highlights the role of media type in shaping these discourses.

Performance Analysis of Embedding Methods for Deep Learning-Based Turkish Sentiment Analysis Models

AbstractThe complex syntactic structure of Turkish text makes sentiment analysis in natural language processing (NLP) a challenging task. Conventional sentiment analysis methods often fail to effectively identify attitudes in Turkish texts, creating an urgent need for more efficient approaches. To fill this need, our study investigates the effectiveness of embedding techniques including pre-trained Turkish models such as Word2Vec, GloVe, and FastText in addition to two character-level embedding methods, namely, character-integer embedding (CIE) and character one-hot encoding embedding (COE), in conjunction with deep learning models specifically long short-term memory (LSTM), convolution neural networks (CNNs), bidirectional LSTM (Bi-LSTM), and hybrid models, for Turkish short-texts sentiment analysis. DL-based models were investigated on two datasets (e.g., an original Twitter (X) dataset and an accessible hotel reviews dataset). In addition to providing an intensive performance analysis of different embedding strategies and assessing their efficacy in dealing with the linguistic intricacies of Turkish, this study proposed a previously unexplored method in Turkish text representation that relies on a character-level one-hot encoding technique. The obtained findings indicate positive progress using a novel approach utilizing a dual-pathway architecture for both character level and word level that constitutes a substantial contribution to the area of natural language processing (NLP), specifically in the context of complex morphological languages. By employing a hybrid strategy that combines character and word levels on Twitter (X) data, the LSTM model obtained anF1 score of$$0.835 \pm 0.005$$0.835±0.005concerning cross-validation while CNN-BiLSTM attained the highestF1 Score (0.8392) using holdout validation. This strategy consistently produced modest improvements across the second public dataset (hotel reviews dataset) by emerging as the runner-up embedding technique in effectiveness, surpassed only by FastText. Findings provide practical recommendations for practitioners on how to effectively use sentiment analysis to make informed decisions by introducing an extensive performance analysis of the use of embedding techniques and deep learning models for sentiment analysis in Turkish texts, which is crucial in the current age of data analysis.

Masculinity and Homosociality in Western Cinema: Taking Close (2022) as a Case Study

Masculinity and homosociality in Western cinema undergo continuous evolution, transitioning from the construction of traditional ideals to contemporary, more diverse expressions, thereby presenting challenges and deconstructions. To contribute to the theorization of masculinity and homosociality in Western films, this study examines Lukas Dhont's Close to analyze how masculinity and homosociality are represented and negotiated to reproduce gender structures and heteronormative ideologies, and offer implications for future film presentations. Employing textual and formal analyses, along with relevant theories of masculinity and homosociality, this paper identifies different forms of masculinities embodied in hierarchical and horizontal homosociality. In doing so, this paper finds that homosociality provides space for some masculinities to assert and solidify their dominance while relegating others to subordinated or marginalized positions. The dynamic shifts in masculinity also correspondingly affect homosocial dynamics. Further exploration is warranted in presenting and constructing positive narratives surrounding non-hegemonic masculinity and horizontal homosociality, so that a diverse, anti-homophobic, and gender equality-supporting masculinity on screen may have the potential to achieve hegemony.

Research on the Content Validity of the English Reading Test in Harbin Senior High School Entrance Examination

The Academic Test for the Junior High School Students, as a kind of academic level examination, plays a crucial role in the selection and streaming of talents at the secondary school level in China. Reading is an essential skill that language learners need to master, and it has a large proportion in the English test of the High School Entrance Examination (HSEE). Therefore, studying the validity of English Reading Test in the HSEE can not only provide the examiner of the HSEE with data references for judging the validity of the reading test, but also provide references for English teachers to innovate teaching methods, as well as cultivating the students' English reading strategies and achieving the overall improvement of the reading ability. Focusing on English Reading Test in Harbin HSEE (2019-2023), this research selects the framework of test characteristics proposed by Bachman and Palmer and analyzes discourse input and expected response to explore the content validity by using textual, quantitative and comparative analyses. The results show that English Reading Test in Harbin HSEE (2019-2023) has a certain degree of content validity and basically meet the requirements of the exam syllabus, but problems still exist. The presentation form of the test text is monotonous, the difficulty gradient is unreasonable and the assessment of reading skills is insufficient.

Assessment Tool of Portfolios: Enhancing Primary School Students' English Reading Proficiency

This study aims to explore the use of portfolios as an assessment tool to enhance primary school students' English reading proficiency. Portfolio assessment offers a more comprehensive and continuous approach compared to traditional assessment methods, which tend to measure students' abilities at a single point in time. By using portfolios, teachers can collect various types of student work that demonstrate their progress and understanding in reading, such as book summaries, text analyses, and personal reflections. This research employs a qualitative method. Data is collected through classroom observations, interviews with teachers, and analysis of students' portfolio documents. The findings indicate that using portfolios as an assessment tool provides several significant benefits. First, portfolios allow for a more holistic evaluation of students' reading abilities, encompassing various aspects such as text comprehension, critical analysis, and reflective skills. Second, portfolios offer students the opportunity to see and recognize their own progress, which can boost their motivation and confidence in learning. Overall, this study concludes that portfolios are an effective assessment tool for enhancing primary school students' reading proficiency in English. The use of portfolios not only aids in more comprehensive assessment but also supports deeper and more sustained development of reading skills. Therefore, it is recommended that schools and educators consider integrating portfolios into their English teaching curriculum.

The role of negative affect in shaping populist support: Converging field evidence from across the globe.

Support for populism has grown substantially during the past 2 decades, a development that has coincided with a marked increase in the experience of negative affect around the world. We use a multimodal, multimethod empirical approach, with data from a diverse set of geographical and political contexts, to investigate the extent to which the rising electoral demand for populism can be explained by negative affect. We demonstrate that negative affect-measured via (a) self-reported emotions in surveys as well as (b) automated text analyses of Twitter data-predicts individual-level populist attitudes in two global surveys (Studies 1a and 1b), longitudinal changes in populist party vote shares at general elections in Europe (Study 2), district-level Brexit voting in the 2016 U.K. referendum (Study 3), and county-level vote shares for Donald Trump in the 2016 and 2020 U.S. presidential elections (Studies 4a and 4b). We find that negative emotions-such as fear and anger as well as more often overlooked low-arousal negative emotions like depression and sadness-are predictive of populist beliefs as well as voting and election results at scale. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Second-level agenda-setting effects of news media and public policy on social media discourse across platforms: immigration during the COVID-19 pandemic in the U.S.

ABSTRACT Integrating intermedia, second-level agenda-setting, and cross-platform research, this study examines the role of news media attention and public policy decisions in driving the salience of issue attributes on multiple social media platforms. We collected three immigration-related datasets (i.e., social media posts, news media headlines, and policy documents covering the first year of the COVID-19 pandemic that also marked the last year of the Trump presidency) and applied manual coding, automated text analysis, and time series modeling to analyze the relationships. The results show that news media set the public agenda across Facebook, Twitter, and YouTube at the second level, with conservative media elevating travel-related discussion and liberal media driving border- and court-rulings-related discussion on those platforms. Compared to news media, the second-level agenda-setting effect of policy announcements was small and platform-specific. Among the three platforms, public discourse on YouTube was the most responsive to media and policy agendas. These findings advance second-level agenda-setting research, demonstrate partisan differences in news media’s impact, reveal the limited role of public policy, and provide a nuanced portrayal of cross-platform public discourse dynamics.

Assessing Illinois companion animal veterinarians' antimicrobial prescription practices and the factors that influence their decisions when treating bacterial infections in dogs and cats.

Judicious antimicrobial use in companion animal practice is critical for maintaining the effectiveness of antimicrobial agents against bacterial infections and reducing the selection of antimicrobial-resistant bacteria. This study aimed to provide insights into companion animal veterinarians' antimicrobial treatment recommendations for common bacterial infections in dogs and cats and describe the factors influencing their prescription choices. An online survey using QualtricsXM® software was administered between September and November 2022 to companion animal veterinarians who were Illinois State Veterinary Medical Association members. Descriptive and text analyses were conducted to assess the participants' responses. A total of 78 surveys were included in the analysis. Skin infections were ranked as the most common bacterial infections for which veterinarians prescribed antimicrobial agents, followed by ear, urinary tract, respiratory, and enteric infections. The severity of clinical symptoms and the results of bacterial culture and susceptibility tests were the most influential factors for veterinarians when making antimicrobial prescription choices. Veterinarians were aware of the current antimicrobial prescription guideline recommendations when prescribing antimicrobials empirically to nine hypothetical scenarios of bacterial infections. According to the results of the text analysis that assessed veterinarians' responses to an open-ended question, regarding their challenges when prescribing antimicrobial agents, the pairwise correlation of word frequencies within each response showed the highest correlations between the words 'owner' and 'compliance', 'administration' and 'route', 'cost' and 'culture', and 'patients' and 'acceptance'. The study results can support animal health stakeholders in the development of antimicrobial stewardship programmes to promote appropriate antimicrobial use and limit the emergence of antimicrobial resistance.

National constitutional courts’ use of the ECJ's precedents: The case study of Czechia

This article interrogates how Member States’ constitutional courts make use of the case-law of the European Court of Justice. Instead of focusing on the extreme instances of open backlash, we put forward a data-driven methodological approach to the study of Europeanization of national constitutional courts. Drawing on the use of automatic text analysis, we suggest that searching for the formal citations of the rulings of the European Court of Justice in the constitutional courts’ case-law contributes to nuanced understanding of their role within the EU. By means of a case study of the Czech Constitutional Court's case-law (more than 71 000 decisions issued between the years 2004 and 2022), we demonstrate a way to explore the so far under-researched area of EU law adjudication where the majority of strategic behaviour might be taking place.

Flowing Body: The Collision of Power Relations, Mind and Body Agency

This article constructs a novel theoretical framework for understanding the body in feminist theory by critically engaging with existing non-binary and non-essentialist views. It advances upon foundational critiques, particularly those by Haraway, Connell, and Butler, integrating the dynamics of body agency, power relations, and mind to propose a more fluid and intersectional approach to body politics. The new framework posits the body as an active participant in its continuous formation, interacting with and reshaping societal norms and expectations. Empirically, the paper analyzed four artworks displayed in art galleries and museums in Lund and Malmö in Sweden as case studies, focusing on how these works represent and shape notions of the female body across various themes including beauty, reproduction, migration, and sexuality. These artworks serve as a medium to explore how corporeal experiences are constructed through the interplay of power relations, mind, and body agency. Through critical analyses of theory and textual analyses of artworks, this paper points out that the body is a language of fluidity, simultaneously influenced and shaped by power relations, the mind, and its agency. Feminism should treat the body more carefully, avoiding biological essentialism or absolute social constructs.

Using Transfer Learning to Realize Low Resource Dungan Language Speech Synthesis

This article presents a transfer-learning-based method to improve the synthesized speech quality of the low-resource Dungan language. This improvement is accomplished by fine-tuning a pre-trained Mandarin acoustic model to a Dungan language acoustic model using a limited Dungan corpus within the Tacotron2+WaveRNN framework. Our method begins with developing a transformer-based Dungan text analyzer capable of generating unit sequences with embedded prosodic information from Dungan sentences. These unit sequences, along with the speech features, provide <unit sequence with prosodic labels, Mel spectrograms> pairs as the input of Tacotron2 to train the acoustic model. Concurrently, we pre-trained a Tacotron2-based Mandarin acoustic model using a large-scale Mandarin corpus. The model is then fine-tuned with a small-scale Dungan speech corpus to derive a Dungan acoustic model that autonomously learns the alignment and mapping of the units to the spectrograms. The resulting spectrograms are converted into waveforms via the WaveRNN vocoder, facilitating the synthesis of high-quality Mandarin or Dungan speech. Both subjective and objective experiments suggest that the proposed transfer learning-based Dungan speech synthesis achieves superior scores compared to models trained only with the Dungan corpus and other methods. Consequently, our method offers a strategy to achieve speech synthesis for low-resource languages by adding prosodic information and leveraging a similar, high-resource language corpus through transfer learning.