Ayurvedic Dosha Classification and Personalized Recommendations using RAG based Chatbot

DOI

10.2991/978-94-6239-616-6_80How to use a DOI?

Keywords

AYUSH; Artificial Intelligence; Dosha Classification; Retrieval-Augmented Generation (RAG); Large Language Model (LLM)

Abstract

Despite advancements in AI-driven healthcare, traditional medical systems such as Ayurveda remain underutilized in modern diagnostic methodologies. Limited integration between ancient Ayurvedic principles and AI-based personalized healthcare reduces the accessibility and effectiveness of holistic wellness solutions. To bridge this gap, an AI-driven system has been developed that combines Ayurvedic diagnostics with machine learning and deep learning techniques for Dosha classification and personalized health recommendations. A multi-stage intelligent system is proposed, beginning with a structured and practitioner-validated questionnaire designed to collect real-time responses, forming a dataset. This dataset is used to train and evaluate a Transformer-based model, which achieved 96.83% classification accuracy, with precision, recall, and F1-scores exceeding 0.95 across most Dosha classes. For interactive response generation, a Retrieval-Augmented Generation (RAG) framework is integrated with a Large Language Model (LLM), enabling a Chatbot interface that delivers dietary, lifestyle, and herbal recommendations. Evaluation of generated responses shows strong performance, with the selected Ayurvedic LLM achieving ROUGE-1 = 0.7931, ROUGE-2 = 0.6071, ROUGE-L = 0.7931, and BERTScore (F1) = 0.9677, demonstrating high semantic accuracy and contextual relevance.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - S. Sendhilkumar
AU  - G. S. Mahalakshmi
AU  - G. Kaushik Hariharan
AU  - V. Logapriya
PY  - 2026
DA  - 2026/03/31
TI  - Ayurvedic Dosha Classification and Personalized Recommendations using RAG based Chatbot
BT  - Proceedings of the International Conference on Artificial Intelligence and Secure Data Analytics (ICAISDA 2025)
PB  - Atlantis Press
SP  - 1105
EP  - 1115
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6239-616-6_80
DO  - 10.2991/978-94-6239-616-6_80
ID  - Sendhilkumar2026
ER  -