Suchergebnisse

Drawing upon the approach of strategic framing, this study investigated how China's state-run media mobilize foreign propaganda machine and use specific patterns to describe the 2019 Hong Kong protests on Twitter. It also shed light on the heterogeneity of both production and reception of the strategic frames used by state media. Structural topic modeling was employed to analyze a large amount of Twitter content (i.e., 14,412 tweets) posted by 13 verified organizational accounts, and six strategic frames were identified as conflicts and violence, calling for stability and order, marginalizing protests, criticizing the West as accomplices, delegitimizing protests, and social and economic disruption. These frames highlighted insider–outsider and causes and consequences as two overarching communication strategies. The results also revealed that the bureaucratic rank of state media and the engagement rate of each tweet were closely associated with the content prevalence of various strategic frames. In addition to enhancing our understanding of the construction of "protest paradigm" against the social media context, these empirical findings uncover the often overlooked mobility and flexibility of China's state media discourse as well as the communication ecology shaped and consolidated by the increasing importance state media communicators attach to online engagement metrics.

Vor dem Hintergrund der Beziehung zwischen gesellschaftlichen Herausforderungen und entsprechenden Beiträgen der Wissenschaft wurde am Beispiel des Themenbereichs Flucht und Migration untersucht, inwiefern die psychologische Forschung jeweils aktuelle gesellschaftliche und politische Fragestellungen aufgreift, ob sie eine evidenzbasierte Grundlage für diese Fragestellungen schafft und welchen Teildisziplinen der Psychologie die Forschungsthemen zugeordnet werden können. Dazu wurden mit Structural Topic Modeling die Abstracts von 4.073 Publikationen aus den deutschsprachigen Ländern, dokumentiert in der psychologischen Referenzdatenbank PSYNDEX und veröffentlicht zwischen 1980 und 2017, analysiert. Es konnten 19 Themen identifiziert werden. Den stärksten zunehmenden Trend zeigten Traumatisierung von Flüchtlingen, transkulturelle Psychotherapie sowie die rehabilitative Behandlung von Patienten mit Migrationshintergrund. Im Bereich der Situation von Ausländern und Gastarbeitern, der Kriminalität von Jugendlichen sowie der sozialen Integration und Akkulturation mit Bezug zur ehemaligen DDR zeigte sich die deutlichste Abnahme der Wahrscheinlichkeit. Das Thema Sprachentwicklung von Migrantenkindern wies die höchste Wahrscheinlichkeit auf, von empirischen Studien behandelt zu werden, das Thema Identitätsentwicklung die höchste Wahrscheinlichkeit von nicht-empirischen Beiträgen. Zusammenfassend wird konstatiert, dass sich im Bereich von Flucht und Migration wesentliche gesellschaftliche und politische Entwicklungen in der psychologischen Fachliteratur widerspiegeln. Hinsichtlich empirischer Beiträge zu den Themen zeigt sich ein gemischtes Bild. Die meisten Themen haben einen klinisch-psychologischen Schwerpunkt, wobei jedoch auch andere Disziplinen vertreten sind. Methodisch kann festgehalten werden, dass der Topic-Model-Ansatz eine hilfreiche Methode für szientometrische Untersuchungen mit großen Textmengen darstellt. ; peerReviewed ; publishedVersion

The Men's Rights Activism (MRA) movement and its sub-movement The Red Pill (TRP), has flourished online, offering support and advice to men who feel their masculinity is being challenged by societal shifts. Whilst some insightful studies have been carried out, the small samples analysed by researchers limits the scope of studies, which is small compared to the large amounts of data that TRP produces. By extracting a significant quantity of content from a prominent MRA website, ReturnOfKings.com (RoK), whose creator is one of the most prominent figures in the manosphere and who has been featured in multiple studies. Research already completed can be expanded upon with topic modelling and neural networked machine learning, computational analysis that is proposed to augment methodologies of open coding by automatically and unbiasedly analysing conceptual clusters. The successes and limitations of this computational methodology shed light on its further uses in sociological research and has answered the question: What can topic modeling demonstrate about the men's rights activism movement's prescriptive masculinity? This methodology not only proved that it could replicate the results of a previous study, but also delivered insights into an increasingly political focus within TRP, and deeper perspectives into the concepts identified within the movement.

A primary challenge for digital library systems when digitizing millions of volumes is to automatically analyze and group the huge document collection by categories while identifying patterns and extracting the main themes. A common method to be leveraged on unlabeled texts is topic modeling. Given the wide range of datasets and evaluation criteria used by researchers, comparing the performance and outputs of existing unsupervised algorithms is a complex task. This paper introduces a domain-based topic modeling evaluation applied to Romanian documents. Several variants of Latent Dirichlet Allocation (LDA) combined with dimensionality reduction techniques were compared to Transformer-based models for topic modeling. Experiments were conducted on two datasets of varying text lengths: abstracts of novels and full-text documents. Evaluations were performed against coherence and silhouette scores, while the validation considered classification and clustering tasks. Results highlighted meaningful topics extracted from both datasets.

Topics describe the main issue discussed in an article, for example: Does an article deal with politics, economics or sports? Field of application/theoretical foundation: In the context of "Agenda Setting", studies analyze which issues are on the public agenda. In the context of "News Values", studies may analyze why some topics are more prominently covered than others. References/combination with other methods of data collection: Many studies combine manual inspection of topics with their automated detection. Quinn et al. (2010) demonstrate for their analyses of legislative speeches how manual inspection may increase the validity of results. Similarly, Hase et al. (2020) use automated content analysis to find and map similar topics for which manual coding is then conducted. Such combinations may contribute to a better and more detailed understanding of topics than automated analyses by themselves. The datasets referred to in the table are described in the following paragraph: Puschmann (2019a) uses New York Times articles (1996-2006, N = 30,862) as well as articles from Die Zeit (2011-2016, N = 377) to identify topics using supervised machine learning. In another tutorial, Puschmann (2019b) uses Sherlock Holmes stories (18th century, N = 12), articles from Die Zeit (2011-2016, N = 377) and debate transcripts (1970-2017, N = 7,897) to apply LDA and structural topic modeling. In her tutorials, Silge (2018a, 2018b) also uses Sherlock Holmes stories (18th century, N = 12) and a news corpus also containing comments (2006-ongoing, N = 100,000). Silge and Robinson (2020) apply LDA topic modeling on news stories by the Associated Press (1992, N = 2,246) as well as books by Dickens, Wells, Verne and Austen (18th century, N = 4). Roberts et al. (2019) use blogposts (2008, N = 13,248) for structural topic modeling. Watanabe and Müller (2019) apply LDA topic modeling on newspaper articles from The Guardian (2016, N = 6,000). Van Atteveldt and Welbers (2019, 2020) use State of the Union speeches (1981-2017, N = 10 and ...

Als zentrales Konstrukt der Kommunikationswissenschaft ist das Thema heute relevanter denn je. Für das Fach besitzt es als Ordnungseinheit der öffentlichen Meinung seit jeher eine besondere Bedeutung, verdankt seinen aktuellen Bedeutungszuwachs über die sozialwissenschaftliche Forschung hinaus aber erst dem Technologiewandel: Mit ihm stieg der Bedarf, exponenziell gewachsene Textmengen in technischen Systemen vollautomatisch ordnen und moderieren zu können. Technisch umgesetzt wird dies unter anderem über Topic-Modeling-Verfahren, welche dem menschlich-intuitiven Themenverständnis ein algorithmisches Pendant zur Seite stellen. Dabei handelt es sich jedoch nicht etwa um eine Automatisierung des klassischen kommunikationswissenschaftlichen Themenkonzepts, sondern vielmehr um eine andere Messart einer anders operationalisierten Themenvariable. Mit der zunehmenden Bedeutung technischer Systeme greift diese algorithmische Themenkonstruktion im Alltag immer mehr Raum, sodass Nutzer*innen bereits heute in der täglichen digitalen Kommunikation (nicht immer bewusst) mit dem Ergebnis einer algorithmischen Modellierung von Themen in Kontakt sind. Für die Kommunikationswissenschaft ergibt sich also eine zweifache Relevanz, sich mit der algorithmischen Themenkonstruktion auseinanderzusetzen: Substanziell ist Topic Modeling als neue Logik der Herstellung und Strukturierung öffentlicher Kommunikation bedeutend; methodisch insofern, dass es dem Fach innovative analytische Zugänge zu bekannten und neuen Forschungsvorhaben ermöglicht. Um dieser zweifachen Relevanz gerecht zu werden, wird die algorithmische Themenkonstruktion in der vorliegenden Arbeit erstens im Gegenstand der Kommunikationswissenschaft begründet und zweitens im Sinne ihrer Fachmethodik verortet. In einer Aufarbeitung der umfassenden interdisziplinären Bezüge wird damit die theoretisch-konzeptionelle Begründungslücke für den Einsatz von Topic Modeling als Methode der empirischen Kommunikationsforschung geschlossen. Die Arbeit macht darauf aufmerksam, wie sich Thematisierungsprozesse in zunehmend algorithmisch moderierter Öffentlichkeit fundamental verändern, wirft ein Schlaglicht auf zugrunde liegende Mechanismen und diskutiert Implikationen für Gesellschaft und Forschung.

As a well-established probabilistic method, topic models seek to uncover latent semantics from plain text. In addition to having textual content, we observe that documents are usually compared in listwise rankings based on their content. For instance, world-wide countries are compared in an international ranking in terms of electricity production based on their national reports. Such document comparisons constitute additional information that reveal documents' relative similarities. Incorporating them into topic modeling could yield comparative topics that help to differentiate and rank documents. Furthermore, based on different comparison criteria, the observed document comparisons usually cover multiple aspects, each expressing a distinct ranked list. For example, a country may be ranked higher in terms of electricity production, but fall behind others in terms of life expectancy or government budget. Each comparison criterion, or aspect, observes a distinct ranking. Considering such multiple aspects of comparisons based on different ranking criteria allows us to derive one set of topics that inform heterogeneous document similarities. We propose a generative topic model aimed at learning topics that are well aligned to multi-aspect listwise comparisons. Experiments on public datasets demonstrate the advantage of the proposed method in jointly modeling topics and ranked lists against baselines comprehensively.

Research has emphasized the limitations of qualitative and quantitative approaches to studying organizational phenomena. For example, in-depth interviews are resource-intensive, while questionnaires with closed-ended questions can only measure predefined constructs. With the recent availability of large textual data sets and increased computational power, text mining has become an attractive method that has the potential to mitigate some of these limitations. Thus, we suggest applying topic modeling, a specific text mining technique, as a new and complementary strategy of inquiry to study organizational phenomena. In particular, we outline the potentials of structural topic modeling for organizational research and provide a step-by-step tutorial on how to apply it. Our application example builds on 428,492 reviews of Fortune 500 companies from the online platform Glassdoor, on which employees can evaluate organizations. We demonstrate how structural topic models allow to inductively identify topics that matter to employees and quantify their relationship with employees' perception of organizational culture. We discuss the advantages and limitations of topic modeling as a research method and outline how future research can apply the technique to study organizational phenomena.