Have you tried Neural Topic Models? Comparative Analysis of Neural and Non-Neural Topic Models with Application to COVID-19 Twitter Data

Bennett, Andrew F.; Misra, Dipendra; Than, Nga

doi:10.48550/arxiv.2105.10165

Cited by 2 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the literature, a large body of research has focused on using Twitter data to understand public discourse about COVID-19 because of its accessibility. The main NLP techniques used, among others, include neural and non-neural topic modeling [67]. Typical existing studies aim to extract useful clinical and health-related information in Twitter data using topic clustering and sentiment analysis [50,53,56,68].…”

Section: Related Workmentioning

confidence: 99%

Agenda-Setting for COVID-19: A Study of Large-Scale Economic News Coverage Using Natural Language Processing

Businger

Dollfus

et al. 2022

Int J Data Sci Anal

View full text Add to dashboard Cite

Over the past two years, organizations and businesses have been forced to constantly adapt and develop effective responses to the challenges of the COVID-19 pandemic. The acuteness, global scale and intense dynamism of the situation make online news and information even more important for making informed management and policy decisions. This paper focuses on the economic impact of the COVID-19 pandemic, using natural language processing (NLP) techniques to examine the news media as the main source of information and agenda-setters of public discourse over an eight-month period. The aim of this study is to understand which economic topics news media focused on alongside the dominant health coverage, which topics did not surface, and how these topics influenced each other and evolved over time and space. To this end, we used an extensive open-source dataset of over 350,000 media articles on non-medical aspects of COVID-19 retrieved from over 60 top-tier business blogs and news sites. We referred to the World Economic Forum’s Strategic Intelligence taxonomy to categorize the articles into a variety of topics. In doing so, we found that in the early days of COVID-19, the news media focused predominantly on reporting new cases, which tended to overshadow other topics, such as the economic impact of the virus. Different independent news sources reported on the same topics, showing a herd behavior of the news media during this global health crisis. However, a temporal analysis of news distribution in relation to its geographic focus showed that the rise in COVID-19 cases was associated with an increase in media coverage of relevant socio-economic topics. This research helps prepare for the prevention of social and economic crises when decision-makers closely monitor news coverage of viruses and related topics in other parts of the world. Thus, monitoring the news landscape on a global scale can support decision-making in social and economic crises. Our analyses point to ways in which this monitoring and issues management can be improved to remain alert to social dynamics and market changes.

show abstract

Section: Related Workmentioning

confidence: 99%

Agenda-Setting for COVID-19: A Study of Large-Scale Economic News Coverage Using Natural Language Processing

Businger

Dollfus

et al. 2022

Int J Data Sci Anal

View full text Add to dashboard Cite

show abstract

How research programs come apart: The example of supersymmetry and the disunity of physics

Gautheron,

Omodei

2023

Quantitative Science Studies

View full text Add to dashboard Cite

According to Peter Galison, the coordination of different “subcultures” within a scientific field happens through local exchanges within “trading zones.” In his view, the workability of such trading zones is not guaranteed, and science is not necessarily driven towards further integration. In this paper, we develop and apply quantitative methods (using semantic, authorship, and citation data from scientific literature), inspired by Galison’s framework, to the case of the disunity of high-energy physics. We give prominence to supersymmetry, a concept that has given rise to several major but distinct research programs in the field, such as the formulation of a consistent theory of quantum gravity or the search for new particles. We show that “theory” and “phenomenology” in high-energy physics should be regarded as distinct theoretical subcultures, between which supersymmetry has helped sustain scientific “trades.” However, as we demonstrate using a topic model, the phenomenological component of supersymmetry research has lost traction and the ability of supersymmetry to tie these subcultures together is now compromised. Our work supports that even fields with an initially strong sentiment of unity may eventually generate diverging research program sand demonstrates the fruitfulness of the notion of trading zones for informing quantitative approaches to scientific pluralism.

show abstract

Have you tried Neural Topic Models? Comparative Analysis of Neural and Non-Neural Topic Models with Application to COVID-19 Twitter Data

Cited by 2 publications

References 20 publications

Agenda-Setting for COVID-19: A Study of Large-Scale Economic News Coverage Using Natural Language Processing

Agenda-Setting for COVID-19: A Study of Large-Scale Economic News Coverage Using Natural Language Processing

How research programs come apart: The example of supersymmetry and the disunity of physics

Contact Info

Product

Resources

About