Structural Topic Model Analysis of Mask-Wearing Issue Using International News Big Data

User feedback on the NHS test & Trace Service during COVID-19: The use of machine learning to analyse free-text data from 37,914 England adults

Objectives

The UK government's approach to the pandemic relies on a test, trace and isolate strategy, mainly implemented via the digital NHS Test & Trace Service. Feedback on user experience is central to the successful development of public-facing Services. As the situation dynamically changes and data accumulate, interpretation of feedback by humans becomes time-consuming and unreliable. The specific objectives were to 1) evaluate a human-in-the-loop machine learning technique based on structural topic modelling in terms of its Service ability in the analysis of vast volumes of free-text data, 2) generate actionable themes that can be used to increase user satisfaction of the Service.

Methods

We evaluated an unsupervised Topic Modelling approach, testing models with 5-40 topics and differing covariates. Two human coders conducted thematic analysis to interpret the topics. We identified a Structural Topic Model with 25 topics and metadata as covariates as the most appropriate for acquiring insights.

Results

Results from analysis of feedback by 37,914 users from May 2020 to March 2021 highlighted issues with the Service falling within three major themes: multiple contacts and incompatible contact method and incompatible contact method, confusion around isolation dates and tracing delays, complex and rigid system.

Conclusions

Structural Topic Modelling coupled with thematic analysis was found to be an effective technique to rapidly acquire user insights. Topic modelling can be a quick and cost-effective method to provide high quality, actionable insights from free-text feedback to optimize public health Services.

News Coverage of Face Masks in Australia During the Early COVID-19 Pandemic: Topic Modeling Study

BACKGROUND

During the COVID-19 pandemic, web-based media coverage of preventative strategies proliferated substantially. News media was constantly informing people about changes in public health policy and practices such as mask-wearing. Hence, exploring news media content on face mask use is useful to analyze dominant topics and their trends.

OBJECTIVE

The aim of the study was to examine news related to face masks as well as to identify related topics and temporal trends in Australian web-based news media during the early COVID-19 pandemic period.

METHODS

Following data collection from the Google News platform, a trend analysis on the mask-related news titles from Australian news publishers was conducted. Then, a latent Dirichlet allocation topic modeling algorithm was applied along with evaluation matrices (quantitative and qualitative measures). Afterward, topic trends were developed and analyzed in the context of mask use during the pandemic.

RESULTS

A total of 2345 face mask-related eligible news titles were collected from January 25, 2020, to January 25, 2021. Mask-related news showed an increasing trend corresponding to increasing COVID-19 cases in Australia. The best-fitted latent Dirichlet allocation model discovered 8 different topics with a coherence score of 0.66 and a perplexity measure of -11.29. The major topics were T1 (mask-related international affairs), T2 (introducing mask mandate in places such as Melbourne and Sydney), and T4 (antimask sentiment). Topic trends revealed that T2 was the most frequent topic in January 2021 (77 news titles), corresponding to the mandatory mask-wearing policy in Sydney.

CONCLUSIONS

This study demonstrated that Australian news media reflected a wide range of community concerns about face masks, peaking as COVID-19 incidence increased. Harnessing the news media platforms for understanding the media agenda and community concerns may assist in effective health communication during a pandemic response.

Constructing big data prevention and control model for public health emergencies in China: A grounded theory study

Big data technology plays an important role in the prevention and control of public health emergencies such as the COVID-19 pandemic. Current studies on model construction, such as SIR infectious disease model, 4R crisis management model, etc., have put forward decision-making suggestions from different perspectives, which also provide a reference basis for the research in this paper. This paper conducts an exploratory study on the construction of a big data prevention and control model for public health emergencies by using the grounded theory, a qualitative research method, with literature, policies, and regulations as research samples, and makes a grounded analysis through three-level coding and saturation test. Main results are as follows: (1) The three elements of data layer, subject layer, and application layer play a prominent role in the digital prevention and control practice of epidemic in China and constitute the basic framework of the “DSA” model. (2) The “DSA” model integrates cross-industry, cross-region, and cross-domain epidemic data into one system framework, effectively solving the disadvantages of fragmentation caused by “information island”. (3) The “DSA” model analyzes the differences in information needs of different subjects during an outbreak and summarizes several collaborative approaches to promote resource sharing and cooperative governance. (4) The “DSA” model analyzes the specific application scenarios of big data technology in different stages of epidemic development, effectively responding to the disconnection between current technological development and realistic needs.