Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.
Who will be affected?
Readers: No access to all 28 journals. We recommend accessing our articles via PubMed Central
Authors: No access to the submission form or your user account.
Reviewers: No access to your user account. Please download manuscripts you are reviewing for offline reading before Wednesday, July 01, 2020 at 7:00 PM.
Editors: No access to your user account to assign reviewers or make decisions.
Copyeditors: No access to user account. Please download manuscripts you are copyediting before Wednesday, July 01, 2020 at 7:00 PM.
Use of Machine Learning Tools in Evidence Synthesis of Tobacco Use among Sexual and Gender Diverse Populations
Shaoying Ma;
Shuning Jiang;
Olivia Yang;
Xuanzhi Zhang;
Yu Fu;
Yusen Zhang;
Aadeeba Kaareen;
Meng Ling;
Jian Chen;
Ce Shang
ABSTRACT
Background:
LGBTQ+ youth and adults use tobacco at a higher rate than the national average in the US. There is an urgent need for synthesizing published evidence to inform tobacco control policies to better serve this priority population.
Objective:
To develop algorithms to curate peer-reviewed articles that study LGBTQ individuals’ tobacco use and that are published at leading tobacco research journals from 2015 to early 2021, and to extract domain-specific textual entities from these articles.
Methods:
Our team built a tobacco research domain-specific semantic database to identify and extract data from articles that studied the LGBTQ+ population. We trained and employed a language model to extract named entities after learning patterns and relationships between words and their context in text.
Results:
Among 2,993 paper abstracts, 33 were identified as relevant to LGBTQ individuals’ tobacco use. We extracted the following information: different groups being studied, within the LGBTQ+ population; geographic locations; product types and characteristics; analytical methods; behavioral outcomes; and policies or interventions.
Conclusions:
Evidence on the impacts of tobacco control policies on the LGBTQ+ population was lacking among the articles from leading tobacco research journals, and our tools have scale-up potentials to be applied to broader LGBTQ+ health literature.
Citation
Please cite as:
Ma S, Jiang S, Yang O, Zhang X, Fu Y, Zhang Y, Kaareen A, Ling M, Chen J, Shang C
Use of Machine Learning Tools in Evidence Synthesis of Tobacco Use Among Sexual and Gender Diverse Populations: Algorithm Development and Validation