The evolution of the ‘Carpathian Basin’ discourse in the Hungarian parliament (1998-2020)

Do radical right-wing populist incumbents strategically adjust their rhetoric in response to declining public support during non-electoral periods? While existing literature has studied populist rhetoric during electoral campaigns, less attention has been paid to how incumbent right-wing populist leaders adapt their communication strategies between elections. This paper examines this question through an analysis of Hungarian Prime Minister Viktor Orbán’s official speeches from 2014 to 2020. Using Structural Topic Modeling (STM), the analysis reveals that Orbán strategically increases his use of specific populist rhetoric – particularly people-centric messages and anti-immigration or nativist rhetoric – when faced with lower approval ratings, even during non-electoral periods. However, contrary to existing theories suggesting populist rhetoric peaks during campaigns, the study finds no significant difference in Orbán’s use of right-wing populist rhetoric between electoral and non-electoral periods. These findings suggest that right-wing populist incumbents remain reliant on populist rhetoric throughout their tenure, likely due to voters’ expectations of rhetorical consistency and the adaptability of right-wing populist rhetoric. This study contributes to our understanding of how right-wing populist leaders strategically deploy rhetoric to maintain public support while in power.

Objectives The purpose of this study was to investigate the effect on learner’s academic emotion by using virtual reality (VR) content in the Earth science class, and to analyze their correlation by analyzing the structural topic model (STM) of the students' academic emotional response.
 Methods For this purpose, a class was conducted using virtual reality (VR) contents in the earth science subject for 54 science high school students. In order to measure changes in academic emotion, pre- and post- academic emotion tests were performed and For the structural topic model (STM) analysis of students' academic emotional responses, students' responses were freely expressed in text. The collected data were analyzed by paired t-test and text mining structural topic model (STM) analysis.
 Results As a result of the pre- and post- paired t-test on academic emotions, the application of virtual reality (VR) content to Earth science subjects had a statistically significant difference in academic emotions and Among academic emotions, there was a statistically significant difference in increasing positive emotions and decreasing negative emotions. In addition, as a result of structural topic model (STM) analysis of academic emotions, when positive emotions were increased, the probability of topic expression for “geological unit learning through VR panorama” was statistically significantly higher, When negative emotions were reduced, it was found that the probability of topic expression for “understanding through experience,” “learning of geological units through VR panorama”, and “learning of geological structures” was statistically significantly improved.
 Conclusions Based on these results, the application of virtual reality (VR) content to earth science subjects can improve positive emotions and reduce negative emotions among students' academic emotions. In addition, as shown in the structural topic model results, the improvement of topic expression probability for geological unit learning due to changes in academic sentiment suggests that it can be used as a teaching/learning method in earth science subject classes.

Content analysis of newspaper coverage of wolf recolonization in France using structural topic modeling

Obtaining and maintaining steady employment can be challenging for people from vulnerable groups. Previous research has focused on the relationship between employer size and employment outcomes for these groups, but the findings have been inconsistent. To clarify this relationship, the current study uses structural topic modeling, a mixed methods research design, to disclose and explain factors behind the association between employer size and labor market outcomes for people from vulnerable groups. The data consist of qualitative interview transcripts concerning the hiring and inclusion of people from vulnerable groups. These were quantitized and analyzed using structural topic modeling. The goals were to investigate topical content and prevalence according to employer size, to provide a comprehensive guide for model estimation and interpretation, and to highlight the wide applicability of this method in social science research. Model estimation resulted in a model with five topics: training, practicalities of the inclusion processes, recruitment, contexts of inclusion, and work demands. The analysis revealed that topical prevalence differed between employers according to size. Thus, these estimated topics can provide evidence as to why the association between employer size and labor market outcomes for vulnerable groups varies across studies––different employers highlight different aspects of work inclusion. The article further demonstrates the strengths and limitations of using structural topic modeling as a mixed methods research design.

Qualitative analysis methods, while crucial for understanding complex factors influencing human behaviors like engagement with digital health technologies, can be conceptually challenging, time-consuming, and resource intensive. To address these challenges, semi-automated text analysis techniques like structural topic modeling (STM) may enhance the efficiency of qualitative analysis. This study investigates how STM can facilitate qualitative data analysis to gain insights into engagement with CareVirtue, a web-based health platform for dementia caregivers. We employed STM on interview data, using the number of CareVirtue journal posts as a covariate to represent caregiver engagement. Our STM model revealed six topics with varying levels of association with engagement. The topics of “recalling positive experiences” and “help from a care network” were associated with high engagement. These topics complement qualitative analysis by providing a deeper understanding of specific topics and demonstrate the potential of quantitative methods, such as STM, to support and augment qualitative analysis.

Open-ended survey questions crucially contribute to researchers’ understandings of respondents’ experiences. However, analyzing open-ended responses using human coders is labor-intensive. Structural topic modeling (STM) is a text mining method that discover topics from textual data. We demonstrate the use of STM to analyze open-ended survey responses to understand how parents coped during the COVID-19 lock-down in Singapore. We administered online surveys to 199 parents in Singapore during the COVID-19 lock-down. To show a STM analysis, we demonstrated a workflow that includes steps in data preprocessing, model estimation, model selection, and model interpretation. An 18-topic model best fit the data based on model diagnostics and researchers’ expertise. Prevalent coping methods described by respondents include “Spousal Support,” “Routines/Schedules,” and “Managing Expectations.” Topic prevalence for some topics varied with respondents’ levels of parenting stress and whether parents were fathers or mothers. STM offers an efficient, valid, and replicable way to analyze textual data such as open-ended survey responses and case notes that can complement researchers’ knowledge and skills. STM can be used as part of a multistage research process or to support other analyses such as clarifying quantitative findings and identifying preliminary themes from qualitative data. Supplemental data for this article is available online at https://doi.org/10.1080/15228835.2022.2036301 .

At the core of the 2008 global financial crisis was a foreclosure crisis in the United States. The federal government focused on responding to the concurrent banking crisis, leaving foreclosure prevention to the states. Despite a nationwide crisis, only some states advanced foreclosure prevention policies. Political theory scholars argue that political ideology and economic interests are the primary drivers of policy outcomes, while discourse scholars argue that themes in the public discourse shape policymaking. In this article, I integrate these literatures to develop and test an account of foreclosure prevention policymaking. To measure discourse, I scrape the text of over 20,000 state-level media publications and inductively code them using Structural Topic Modeling. Using event history analysis, I examine the relationship between discourse themes, political factors, and the timing of foreclosure prevention policymaking by state legislatures. I find that states where a “markets” theme was more prevalent in the foreclosure discourse were less likely to advance foreclosure prevention policies, whereas states with discourse focused on “intervention” were more likely to do so. Results also corroborate previous scholarship showing that political ideology and special interest group activity impacted these policy outcomes.

This study explored user satisfaction with mobile payments by applying a novel structural topic model. Specifically, we collected 17,927 online reviews of a specific mobile payment (i.e., PayPal). Then, we employed a structural topic model to investigate the relationship between the attributes extracted from online reviews and user satisfaction with mobile payment. Consequently, we discovered that “lack of reliability” and “poor customer service” tend to appear in negative reviews. Whereas, the terms “convenience,” “user-friendly interface,” “simple process,” and “secure system” tend to appear in positive reviews. On the basis of information system success theory, we categorized the topics “convenience,” “user-friendly interface,” and “simple process,” as system quality. In addition, “poor customer service” was categorized as service quality. Furthermore, based on the previous studies of trust and security, “lack of reliability” and “secure system” were categorized as trust and security, respectively. These outcomes indicate that users are satisfied when they perceive that system quality and security of specific mobile payments are great. On the contrary, users are dissatisfied when they feel that service quality and reliability of specific mobile payments is lacking. Overall, our research implies that a novel structural topic model is an effective method to explore mobile payment user experience.

The Querdenken movement, the leading force behind German corona protests, is suspected of being a gateway to far-right attitudes due to radicalizing inward-oriented communication on Telegram. To investigate potential connections of this movement to the far right and alternative media—and to explore key topics of the Querdenken network over time—we analyzed 6,294,955 messages from 578 public Telegram channels via network analysis and structural topic modeling. This analysis revealed that Querdenken’s subcommunities preferably forward content from far-right and QAnon communities, while far-right and conspiracy theorist alternative media channels act as content distributors for the movement. Four main topics appeared in the Querdenken network with varying prevalence over time and across different communities: promotion, QAnon, right-wing populism, and COVID-19 conspiracy theories. Our results highlight potential directions for future research and practical implications, for example, that political decision makers should account for the increasing influence of the QAnon movement on Querdenken mobilizers’ Telegram activity.

Étude prospective de la résistance aux azolés d’Aspergillus fumigatus

BackgroundHealth care organizations are collecting increasing volumes of clinical text data. Topic models are a class of unsupervised machine learning algorithms for discovering latent thematic patterns in these large unstructured document collections.ObjectiveWe aimed to comparatively evaluate several methods for estimating temporal topic models using clinical notes obtained from primary care electronic medical records from Ontario, Canada.MethodsWe used a retrospective closed cohort design. The study spanned from January 01, 2011, through December 31, 2015, discretized into 20 quarterly periods. Patients were included in the study if they generated at least 1 primary care clinical note in each of the 20 quarterly periods. These patients represented a unique cohort of individuals engaging in high-frequency use of the primary care system. The following temporal topic modeling algorithms were fitted to the clinical note corpus: nonnegative matrix factorization, latent Dirichlet allocation, the structural topic model, and the BERTopic model.ResultsTemporal topic models consistently identified latent topical patterns in the clinical note corpus. The learned topical bases identified meaningful activities conducted by the primary health care system. Latent topics displaying near-constant temporal dynamics were consistently estimated across models (eg, pain, hypertension, diabetes, sleep, mood, anxiety, and depression). Several topics displayed predictable seasonal patterns over the study period (eg, respiratory disease and influenza immunization programs).ConclusionsNonnegative matrix factorization, latent Dirichlet allocation, structural topic model, and BERTopic are based on different underlying statistical frameworks (eg, linear algebra and optimization, Bayesian graphical models, and neural embeddings), require tuning unique hyperparameters (optimizers, priors, etc), and have distinct computational requirements (data structures, computational hardware, etc). Despite the heterogeneity in statistical methodology, the learned latent topical summarizations and their temporal evolution over the study period were consistently estimated. Temporal topic models represent an interesting class of models for characterizing and monitoring the primary health care system.

In the English-speaking world Ármin Vámbéry is known as a traveler in Central Asia and a student of Turkic cultures and languages. In his native Hungary he is also known for his disagreement with linguists who believed that Hungarian belonged to the Ugric branch of the Finno-Ugric languages—a part of the Uralic linguistic family. Rather than accepting this theory, Vámbéry contended that Hungarian was largely a Turkic language that belonged more to the Altaic family. Few people know that Vámbéry also expressed strong opinions about the genesis of the Hungarian nation. The most important aspect of Vámbéry’s theory about Hungarian origins is the thesis that Hungarian ethnogenesis took place—beginning with late Roman times or even earlier—in the Carpathian Basin. A corollary of this proposition is that the nomadic tribes that conquered the Carpathian Basin at the end of the ninth century were Turkic peoples who were few in numbers and were assimilated by the region’s autochthonous—and by then Hungarian-speaking—population. This paper outlines Vámbéry’s arguments and describes to what extent research on this subject in the century since Vámbéry’s death has confirmed or contradicted his unconventional ideas.

The past few decades have seen the rapid development of topic modeling. So far, research has been more concerned with determining the ideal number of topics or meaningful topic clustering words than with applying topic modeling techniques to evaluate linguistic theories. This study proposes the Structural Topic Model (STM)-led framework to facilitate the interpretation of topic modeling results and standardize text analysis. STM encompasses various model training mechanisms, thereby requiring systematic designs to properly combine language studies. “Structural” in STM refers to the inclusion of metadata structure. Unlike the corpus-based keyness approach, STM can capture contextual cues and meta-information for the interpretation of topical results. Besides, STM can make cross-corpora comparisons via topical contrast, a challenging task for corpus-driven related models such as the Biterm Topic Model (BTM). Stylistic variations in song lyrics are taken as an illustration to show how to use the suggested framework to delve into the linguistic theory proposed by Pennebaker (2013). The topical model and iterable model in the proposed paradigm can clarify how pronouns affect style distinction. We believe the proposed STM-led framework can shed light on text analysis by conducting a reproducible cross-corpora comparison on short texts.

Describing the patient experience from Yelp reviews of community pharmacies

BackgroundStressors for health care workers (HCWs) during the COVID-19 pandemic have been manifold, with high levels of depression and anxiety alongside gaps in care. Identifying the factors most tied to HCWs’ psychological challenges is crucial to addressing HCWs’ mental health needs effectively, now and for future large-scale events.ObjectiveIn this study, we used natural language processing methods to examine deidentified psychotherapy transcripts from telemedicine treatment during the initial wave of COVID-19 in the United States. Psychotherapy was delivered by licensed therapists while HCWs were managing increased clinical demands and elevated hospitalization rates, in addition to population-level social distancing measures and infection risks. Our goal was to identify specific concerns emerging in treatment for HCWs and to compare differences with matched non-HCW patients from the general population.MethodsWe conducted a case-control study with a sample of 820 HCWs and 820 non-HCW matched controls who received digitally delivered psychotherapy in 49 US states in the spring of 2020 during the first US wave of the COVID-19 pandemic. Depression was measured during the initial assessment using the Patient Health Questionnaire-9, and anxiety was measured using the General Anxiety Disorder-7 questionnaire. Structural topic models (STMs) were used to determine treatment topics from deidentified transcripts from the first 3 weeks of treatment. STM effect estimators were also used to examine topic prevalence in patients with moderate to severe anxiety and depression.ResultsThe median treatment enrollment date was April 15, 2020 (IQR March 31 to April 27, 2020) for HCWs and April 19, 2020 (IQR April 5 to April 27, 2020) for matched controls. STM analysis of deidentified transcripts identified 4 treatment topics centered on health care and 5 on mental health for HCWs. For controls, 3 STM topics on pandemic-related disruptions and 5 on mental health were identified. Several STM treatment topics were significantly associated with moderate to severe anxiety and depression, including working on the hospital unit (topic prevalence 0.035, 95% CI 0.022-0.048; P<.001), mood disturbances (prevalence 0.014, 95% CI 0.002-0.026; P=.03), and sleep disturbances (prevalence 0.016, 95% CI 0.002-0.030; P=.02). No significant associations emerged between pandemic-related topics and moderate to severe anxiety and depression for non-HCW controls.ConclusionsThe study provides large-scale quantitative evidence that during the initial wave of the COVID-19 pandemic, HCWs faced unique work-related challenges and stressors associated with anxiety and depression, which required dedicated treatment efforts. The study further demonstrates how natural language processing methods have the potential to surface clinically relevant markers of distress while preserving patient privacy.