Enhancing Medical Information Retrieval using RAG with Hybrid Vector Search and Query Expansion
Volume 19, Issue 1, 2025
Download| Author(s): |
Irtiza Hussain* NED University of Engineering and Technolog, Karachi, Pakistan, irtizahussain221@gmail.com |
|---|---|
| Abstract | The rise of large language models (LLMs) has improved people’s access to medical information. These LLMs, however, lack the domain-specific knowledge essential to answering more specialized questions. To plug this knowledge gap, this paper explores the use of retrieval-augmented generation. A comparative analysis of four different retrieval strategies — dense vector search, dense vector search with query expansion, hybrid vector search, and hybrid vector search with query expansion — was carried out and a combination of hybrid vector search with query expansion was found to be the most effective retrieval strategy. This framework will help efficient and cost effective information retrieval for chatbot used in the medical sector. At the same time, because of its generic nature, this proposed combination can be used for applications in other fields. |
| Keywords | Retrieval-Augmented Generation, Hybrid Vector Search, Query Expansion, LLMs, Dense Vectors, Sparse Vectors |
| Year | 2025 |
| Volume | 19 |
| Issue | 1 |
| Type | Research paper, manuscript, article |
| Journal Name | Journal of Information & Communication Technology | Publisher Name | ILMA University | Jel Classification | - | DOI | - | ISSN no (E, Electronic) | 2075-7239 | ISSN no (P, Print) | 2415-0169 | Country | Pakistan | City | Karachi | Institution Type | University | Journal Type | Open Access | Manuscript Processing | Blind Peer Reviewed | Format | Paper Link | https://jict.ilmauniversity.edu.pk/journal/jict/19.1/4.pdf | Page | 14-20 |