|
Abstract
|
BACKGROUND: In most industrialized societies, there are regulations, inspections, insurance, and legal options to support workers who suffer from injury, disease, or death in relation to their work; in practice these resources are imperfect and or even being unavailable due to workplace or employer obstruction. Thus, limitations exist to identify unmet needs in occupational safety and health information.
OBJECTIVE: This study was aimed to explore hidden issues in occupational accidents from social network services (SNS) data using topic modeling.
METHODS: Using the results of a Google search for the phrase "occupational/industrial accident" and "occupational diseases," a total of 145 websites were selected. From among these websites, we collected 15,244 documents on occupational accidents-related queries between 2002 and 2018. To transform unstructured text into structure data, natural language processing of the Korean language was conducted. We performed the Latent Dirichlet allocation (LDA) as a topic model using Python library. A time-series linear regression analysis was also conducted to identify yearly trends for the given documents.
RESULTS: Results of LDA model showed 14 topics with three themes: "Workers' compensation benefits" (Theme 1), "Illicit agreements with the employer" (Theme 2), and "Fatal and non-fatal injuries and vulnerable workers" (Theme 3). Theme 1 was the largest cluster (52.2%) of the collected documents and included keywords related to workers' compensation (i.e., company, occupational injury, insurance, accident, approval, and compensation) and keywords describing the specific compensation benefits such as medical expense benefits, temporary incapacity benefits, and disability benefits. In the yearly trend, Theme 1 gradually decreased, but other themes showed an overall increasing pattern. Certain queries (i.e., musculoskeletal system, critical care, and foreign workers) had no significant variation in the number of queries.
CONCLUSIONS: We conducted LDA analysis with SNS data of occupational accidents-related queries and discovered that the primary concerns of workers posting on occupational injuries and diseases were workers' compensation benefits, fatal and non-fatal injuries and vulnerable workers, and illicit agreements with the employer.. While traditional systems focus mainly on quantitative monitoring of occupational accidents, qualitative aspects formulated by topic modeling from unstructured SNS queries may be valuable for tackling inequality and improving occupational health and safety.
CLINICALTRIAL:
Language: en |