Dataset Information

Inferring global-scale temporal latent topics from news reports to predict public health interventions for COVID-19.

ABSTRACT: The COVID-19 pandemic has highlighted the importance of non-pharmacological interventions (NPIs) for controlling epidemics of emerging infectious diseases. Despite their importance, NPIs have been monitored mainly through the manual efforts of volunteers. This approach hinders measurement of the NPI effectiveness and development of evidence to guide their use to control the global pandemic. We present EpiTopics, a machine learning approach to support automation of NPI prediction and monitoring at both the document level and country level by mining the vast amount of unlabeled news reports on COVID-19. EpiTopics uses a 3-stage, transfer-learning algorithm to classify documents according to NPI categories, relying on topic modeling to support result interpretation. We identified 25 interpretable topics under 4 distinct and coherent COVID-related themes. Importantly, the use of these topics resulted in significant improvements over alternative automated methods in predicting the NPIs in labeled documents and in predicting country-level NPIs for 42 countries.

SUBMITTER: Wen Z

PROVIDER: S-EPMC8805211 | biostudies-literature | 2022 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Inferring global-scale temporal latent topics from news reports to predict public health interventions for COVID-19.

Wen Zhi Z Powell Guido G Chafi Imane I Buckeridge David L DL Li Yue Y

Patterns (New York, N.Y.) 20220201 3

The COVID-19 pandemic has highlighted the importance of non-pharmacological interventions (NPIs) for controlling epidemics of emerging infectious diseases. Despite their importance, NPIs have been monitored mainly through the manual efforts of volunteers. This approach hinders measurement of the NPI effectiveness and development of evidence to guide their use to control the global pandemic. We present EpiTopics, a machine learning approach to support automation of NPI prediction and monitoring a ...[more]

PMID: 35128492

Similar Datasets

Project description:BackgroundRapid identification of subject experts for medical topics helps in improving the implementation of discoveries by speeding the time to market drugs and aiding in clinical trial recruitment, etc. Identifying such people who influence opinion through social network analysis is gaining prominence. In this work, we explore how to combine named entity recognition from unstructured news articles with social network analysis to discover opinion leaders for a given medical topic.MethodsWe employed a Conditional Random Field algorithm to extract three categories of entities from health-related new articles: Person, Organization and Location. We used the latter two to disambiguate polysemy and synonymy for the person names, used simple rules to identify the subject experts, and then applied social network analysis techniques to discover the opinion leaders among them based on their media presence. A network was created by linking each pair of subject experts who are mentioned together in an article. The social network analysis metrics (including centrality metrics such as Betweenness, Closeness, Degree and Eigenvector) are used for ranking the subject experts based on their power in information flow.ResultsWe extracted 734,204 person mentions from 147,528 news articles related to obesity from January 1, 2007 through July 22, 2010. Of these, 147,879 mentions have been marked as subject experts. The F-score of extracting person names is 88.5%. More than 80% of the subject experts who rank among top 20 in at least one of the metrics could be considered as opinion leaders in obesity.ConclusionThe analysis of the network of subject experts with media presence revealed that an opinion leader might have fewer mentions in the news articles, but a high network centrality measure and vice-versa. Betweenness, Closeness and Degree centrality measures were shown to supplement frequency counts in the task of finding subject experts. Further, opinion leaders missed in scientific publication network analysis could be retrieved from news articles.

Project description:The aim of this study is to reveal a robust correlation between the amount of attention international journalism devotes to scientific papers and the amount of attention scientific journals devote to the respective topics. Using a Mainstream-Media-Score (MSM) ≥ 100 (which we regard as an indicator for news media attention) from the altmetrics provider Altmetric, we link 983 research articles with 185,166 thematically similar articles from the PubMed database (which we use to operationalize attention from scientific journals). The method we use is to test whether there is a concomitant increase in scientific attention after a research article has received popular media coverage. To do so, we compare the quotient of the number of thematically similar articles published in scientific journals during the period before and after the publication of an MSM ≥ 100 article. Our main result shows that in 59 percent of cases, more thematically similar articles were published in scientific journals after a scientific paper received noteworthy news media coverage than before (p < 0.01). In this context, we neither found significant differences between various types of scientific journal (p = 0.3) nor between scientific papers that were originally published in renowned opinion-leading journals or in less renowned, non-opinion-leading journals (p = 0.1). Our findings indicate a robust correlation between the choice of topics in the mass media and in research. However, our study cannot clarify whether this correlation occurs because researchers and/or scientific journals are oriented towards public relevance (publicity effect) or whether the correlation is due to the parallelism of relevance attributions in quality journalism and research (earmark hypothesis). We infer that topics of social relevance are (more) likely to be picked up by popular media as well as by scientific journals. Altogether, our study contributes new empirical findings to the relationship between topic selection in journalism and in research.

Dataset Information

Inferring global-scale temporal latent topics from news reports to predict public health interventions for COVID-19.

Publications

Inferring global-scale temporal latent topics from news reports to predict public health interventions for COVID-19.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets