Harnessing Latent Dirichlet Allocation-based topic modeling to unveil the focus areas in Ayush healthcare in India

The exponential growth of unstructured data has presented new opportunities for leveraging computational techniques to uncover meaningful insights in diverse domains, including healthcare. In the context of India's pluralistic medical ecosystem, the Ministry of Ayush plays a central role in shaping policies related to traditional systems such as Ayurveda, Yoga, Naturopathy, Unani, Siddha, and Homeopathy. Despite the increasing prominence of AYUSH, especially post-COVID-19, there has been limited scholarly engagement with its policy evolution through a data-driven lens. This study applies Latent Dirichlet Allocation (LDA), a topic modeling technique, to analyze annual reports published by the Ministry of Ayush from 2012 to 2023. The research aims to (i) identify recent focus areas in alternate healthcare policy, (ii) track temporal shifts in strategic priorities, and (iii) uncover underrepresented themes in institutional discourse. The findings reveal evolving trends such as increased emphasis on research, integration with mainstream health systems, digital initiatives, and global promotion. The study identifies four critical dimensions that remain underrepresented within Ayush policy discourse: the positioning of Ayush philosophy as a frontline public health framework, the articulation of an Ayush-centric health system paradigm, the programmatic integration of mental health, and the systematic incorporation of artificial intelligence beyond foundational digital health initiatives, thereby identifying areas requiring greater strategic attention and reform. By introducing topic modeling into Indian healthcare policy research, this paper fills a key methodological gap and offers empirical insights into the policy trajectory of Ayush. To the best of the authors' knowledge, this is the first application of LDA to Ayush's policy literature.