Authors : Shivam Kumar; Chetan D. Kachroo; Charnpreet Kaur; Aditya Sharma Vats; Bilal Ahmad; Prakhar Kumar Singh

Volume/Issue : Volume 8 - 2023, Issue 11 - November

Google Scholar : https://tinyurl.com/28ev6weh

Scribd : https://tinyurl.com/bduwt4ez

DOI : https://doi.org/10.5281/zenodo.10159700

Abstract : This research paper introduces a cutting-edge healthcare chatbot that harnesses the synergy of Llama2, Faiss, and Hugging Face embeddings to optimize responses to intricate medical inquiries. Leveraging a meticulously curated training corpus of medical literature, this chatbot significantly augments its semantic understanding and responsiveness. The integration of Llama2 bolsters the chatbot’s contextual comprehension, while Faiss enables expedited, similarity-based information retrieval from an extensive library of medical texts. Hugging Face embed-dings facilitate contextually coherent response generation. The results affirm substantial enhancements in the chatbot’s efficacy in delivering technically informed and contextually precise medical responses. This promising innovation offers a powerful tool for disseminating validated medical knowledge, serving as an invaluable resource for healthcare professionals and patients alike.

Keywords : Healthcare Chatbot, Llama2, Faiss, Hugging Face Embeddings.