The Application of RAG Technology in Traditional Chinese Medicine

DOI

10.2991/978-94-6463-512-6_43How to use a DOI?

Keywords

Large language model; retrieval augmented generation; Chinese tradition medicine

Abstract

The inheritance and development of traditional Chinese medicine are facing limitations and significant challenges in today's society. This article combines the Large Language Model (LLM) model with the Retrieval Augmented Generation (RAG) model to address this issue, helping medical students to quickly retrieve highly specialized knowledge and to some extent, assisting in the modernization and inheritance of traditional Chinese medicine. Taking the Compendium of Materia Medica as an example, this article divides the text into blocks and vectorizes them, splitting the initial text into different blocks, and then selecting an optimization model to embed the text blocks. Afterwards, an index is established for the text, and after retrieval, filters and reordering techniques are used to further refine the search results. Next, this article constructs a chat engine to provide chat logic for the RAG system, using query compression technology to solve problems related to subsequent support and pronoun reference, and helping the chat engine to consider the context of the conversation. Finally, this article sends the context of the unlocked results block by block to LLM to optimize the answer, while summarizing the retrieved context, adapting prompts, and generating multiple answers based on different context blocks, connecting and summarizing them. Based on the relevant experimental results, the model's text matching degree is relatively high. At the same time, the model's recall rate and text accuracy rate are acceptable, indicating that prompt word engineering is still needed to improve the accuracy of knowledge retrieval in traditional Chinese medicine.

Copyright

© 2024 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  - Yufeng Liu
PY  - 2024
DA  - 2024/09/23
TI  - The Application of RAG Technology in Traditional Chinese Medicine
BT  - Proceedings of the 2024 International Conference on Artificial Intelligence and Communication (ICAIC 2024)
PB  - Atlantis Press
SP  - 402
EP  - 408
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-512-6_43
DO  - 10.2991/978-94-6463-512-6_43
ID  - Liu2024
ER  -