scholarly journals Health Communication for Hospice Care through Chinese Media: Digital Topic Modeling Approach (Preprint)

10.2196/29375 ◽  
2021 ◽  
Author(s):  
Qian Liu ◽  
Zequan Zheng ◽  
Jingsen Chen ◽  
Winghei Tsang ◽  
Jin Shan ◽  
...  
Keyword(s):  
Health Communication ◽  
Topic Modeling ◽  
Hospice Care ◽  
Chinese Media ◽  
Modeling Approach

Health Communication for Hospice Care through Chinese media: Digital Topic Modeling Approach (Preprint)

2021 ◽  
Author(s):  
Qian Liu ◽  
Zequan Zheng ◽  
Jingsen Chen ◽  
Winghei Tsang ◽  
Jin Shan ◽  
...  
Keyword(s):  
Mass Media ◽  
Health Communication ◽  
Topic Modeling ◽  
Health Care Services ◽  
Hospice Care ◽  
Patient Treatment ◽  
News Reports ◽  
Care Services ◽  
Time Period ◽  
Second Period

BACKGROUND Hospice care, a type of end-of-life care provided for dying patients and their families, has been rooted in China since the 1980s. It can improve receivers’ quality of life as well as ease their economic burden. The Chinese mass media have continued to actively dispel misconceptions of hospice care and deliver the latest information to citizens. OBJECTIVE This study aimed to retrieve and analyze news reports on hospice care to gain insight into whether any differences exist in delivered heath information as time went by and the role the mass media played in health communication in recent years. METHODS We searched the Huike (WiseSearch) database for related news from Chinese mass media between 2014 and 2019. We set January 1, 2014 to December 31, 2016 as the first time period and January 1, 2017 to December 31, 2019 as the second time period. Python was used to complete the data cleaning process. We determined appropriate topic numbers for these two periods based on coherence score and applied the latent Dirichlet allocation topic modeling. Keywords of each topic and corresponding topics’ names were then generated. The topics were plotted into different circles and their distances on the two-dimensional plane was represented by multidimensional scaling. RESULTS After removing the duplicated and irrelevant news articles, we obtained a total of 2227 articles. We chose eight as the suitable topic number for both time periods and generated topics’ name and their keywords. The top three most reported topics in the first period were patient treatment, hospice care stories, and development of health care services and health insurance, accounting for 18.68% (n = 178), 16.58% (n = 158), and 14.17% (n = 135) of the collected reports, respectively. The top three most reported topics in the second period were hospice care stories, patient treatment, and development of health care services, accounting for 15.62% (n = 199), 15.38 (n = 15.38), and 14.27% (n = 182), respectively. CONCLUSIONS Topic modeling of news reports gives us a better understanding of patterns of health communication about hospice care by mass media. Chinese mass media frequently reported on hospice care in April due to a traditional Chinese festival. An increase in coverage in the second period was observed. These two periods share six similar topics, among which patient treatment outstrips hospice care stories as the most-reported topic in the second period, showing the humanistic spirit behind the reports. We suggest stakeholders cooperate with the mass media when planning to update policies.

scholarly journals Modeling Human Factors Topics in Aviation Reports

2019 ◽  
Vol 63 (1) ◽  
pp. 126-130
Author(s):  
Beth Lyall-Wilson ◽  
Nicolas Kim ◽  
Elizabeth Hohman
Keyword(s):  
Human Factors ◽  
Topic Modeling ◽  
Domain Knowledge ◽  
Aviation Safety ◽  
Subject Matter Experts ◽  
Text Data ◽  
Modeling Approach ◽  
Modeling Process ◽  
Manual Review ◽  
Initial Extraction

This paper describes the development and new application of a text modeling process for identifying human factors topics, such as fatigue, workload, and distraction in aviation safety reports. Current approaches to identifying human factors topic representations in text data rely on manual review from subject matter experts. The implementation of a semi-supervised text modeling method overcomes the need for lengthy manual review through an initial extraction of pre-defined human factors topics, freeing time for focus on analyzing the information. This modeling approach allows analysts to use keywords to define topics of interest up front and influence the convergence of the model toward a result that reflects them, which provides an advantage over classic topic modeling approaches where domain knowledge is not integrated into the generation of derived topics. This paper includes a description of the modeling approach and rationale, data used, evaluation methods, challenges, and suggestions for future applications.

scholarly journals What Does Personality Mean in the Context of Mental Health? A Topic Modeling Approach Based on Abstracts Published in Pubmed Over the Last 5 Years

2020 ◽  
Vol 10 ◽  
Author(s):  
Raffaele Sperandeo ◽  
Giovanni Messina ◽  
Daniela Iennaco ◽  
Francesco Sessa ◽  
Vincenzo Russo ◽  
...  
Keyword(s):  
Mental Health ◽  
Topic Modeling ◽  
Modeling Approach

An Evaluation of Patient Safety Event Report Categories Using Unsupervised Topic Modeling

2015 ◽  
Vol 54 (04) ◽  
pp. 338-345 ◽  
Author(s):  
A. Fong ◽  
R. Ratwani
Keyword(s):  
Patient Safety ◽  
Natural Language Processing ◽  
Natural Language ◽  
Language Processing ◽  
Topic Modeling ◽  
Free Text ◽  
Event Data ◽  
Event Type ◽  
Modeling Approach ◽  
Safety Event

SummaryObjective: Patient safety event data repositories have the potential to dramatically improve safety if analyzed and leveraged appropriately. These safety event reports often consist of both structured data, such as general event type categories, and unstructured data, such as free text descriptions of the event. Analyzing these data, particularly the rich free text narratives, can be challenging, especially with tens of thousands of reports. To overcome the resource intensive manual review process of the free text descriptions, we demonstrate the effectiveness of using an unsupervised natural language processing approach.Methods: An unsupervised natural language processing technique, called topic modeling, was applied to a large repository of patient safety event data to identify topics, or themes, from the free text descriptions of the data. Entropy measures were used to evaluate and compare these topics to the general event type categories that were originally assigned by the event reporter.Results: Measures of entropy demonstrated that some topics generated from the un-supervised modeling approach aligned with the clinical general event type categories that were originally selected by the individual entering the report. Importantly, several new latent topics emerged that were not originally identified. The new topics provide additional insights into the patient safety event data that would not otherwise easily be detected.Conclusion: The topic modeling approach provides a method to identify topics or themes that may not be immediately apparent and has the potential to allow for automatic reclassification of events that are ambiguously classified by the event reporter.

Drawing openness to experience from user generated contents: An interpretable data-driven topic modeling approach

2020 ◽  
Vol 144 ◽  
pp. 113073 ◽  
Author(s):  
Yishi Zhang ◽  
Haiying Wei ◽  
Yaxuan Ran ◽  
Yang Deng ◽  
Dan Liu
Keyword(s):  
Topic Modeling ◽  
Openness To Experience ◽  
Data Driven ◽  
Modeling Approach

scholarly journals On Building an Interpretable Topic Modeling Approach for the Urdu Language

2020 ◽  
Author(s):  
Zarmeen Nasim
Keyword(s):  
Feature Extraction ◽  
Deep Learning ◽  
Topic Modeling ◽  
Language Modeling ◽  
Low Resource ◽  
Modeling Approach ◽  
Part Of Speech ◽  
Modeling Techniques ◽  
Pos Tagger

This research is an endeavor to combine deep-learning-based language modeling with classical topic modeling techniques to produce interpretable topics for a given set of documents in Urdu, a low resource language. The existing topic modeling techniques produce a collection of words, often un-interpretable, as suggested topics without integrat-ing them into a semantically correct phrase/sentence. The proposed approach would first build an accurate Part of Speech (POS) tagger for the Urdu Language using a publicly available corpus of many million sentences. Using semanti-cally rich feature extraction approaches including Word2Vec and BERT, the proposed approach, in the next step, would experiment with different clus-tering and topic modeling techniques to produce a list of potential topics for a given set of documents. Finally, this list of topics would be sent to a labeler module to produce syntactically correct phrases that will represent interpretable topics.

Sign in / Sign up

Export Citation Format

Share Document