Enterprise AI Analysis

Research Hotspots and Topic Evolution in the Field of Education and Public Health: An LDA-Based Text Mining Analysis

This study employs an optimized LDA topic model and text mining technology to conduct an intelligent analysis of research literature in the interdisciplinary field of education and public health. It systematically reveals the research hotspots, structural relationships, and development trends in China's education and public health domain from 2020 to 2024, while validating the effectiveness of computational linguistics methods in this field of research. Methods: Using a sample of 335 relevant research articles indexed in Peking University Core and CSSCI journals from the CNKI database, the text data underwent comprehensive preprocessing. This included automated data cleaning (removal of punctuation, numbers, and non-textual symbols), tokenization using the jieba segmentation module, and linguistic normalization steps such as stop-word removal (based on customized lists from Harbin Institute of Technology and Baidu) and synonym merging (e.g., unifying "Health Education" and "Health Promotion"). The optimized LDA topic model was utilized for topic identification. Combined with computational linguistics coherence metrics and pyLDAvis interactive visualization technology, the optimal number of topics was determined. Subsequently, the topic content, evolutionary paths, and intrinsic relationships were analyzed.Results: The study identified three core themes: Health Literacy and Curriculum Integration (Topic 0), School Health Policy and Health Promotion (Topic 1), and Mental Health and Behavioral Intervention (Topic 2). Word cloud and high-frequency word analysis based on natural language processing further indicated that "Digital Health," "Campus Public Health," and "Health Equity" have emerged as the most prominent frontier issues. The thematic evolution trend shows that the research focus has systematically deepened from early infectious disease emergency education towards the cultivation of routine health literacy and the development of an ecological health sup-port system.Conclusion: This research reveals a complete research framework in the field of education and public health characterized by "goal orientation, policy promotion, technology empowerment, and ecological support." It also demonstrates the effectiveness and practicality of computational text analysis methods in this domain. Future efforts will further deepen the development of health literacy-oriented curriculum evaluation systems, human-computer collaborative health education models, and the construction of a cross-sectoral collaborative public health education ecology. Con-currently, the application of artificial intelligence technology in education and public health research should be strengthened to promote the integrated innovation and high-quality development of educational practices and public health initiatives.