Topic Modeling Analysis of Children's Food Safety Management Using BigKinds News Big Data: Comparing the Implementation Times of the Comprehensive Plan for Children's Dietary Safety Management.

As digital technologies and food environments evolve, ensuring children's food safety has become a pressing public health priority. This study examines how the policy discourse on children's dietary safety in Korea has shifted over time by applying Latent Dirichlet Allocation (LDA) topic modeling to news articles from 2010 to 2024. Using a large-scale news database (BigKinds), the analysis identifies seven key themes that have emerged across five phases of the national Comprehensive Plans for Safety Management of Children's Dietary Life. These include experiential education, data-driven policy approaches, safety-focused meal management, healthy dietary environments, nutritional support for children's growth, customized safety education, and private-sector initiatives. A significant increase in digital keywords-such as "big data" and "artificial intelligence"-highlights a growing emphasis on data-oriented policy tools. By capturing the evolving language and priorities in food safety policy, this study provides new insights into the digital transformation of public health governance and offers practical implications for adaptive and technology-informed policy design.