Open Access Open Access  Restricted Access Subscription Access

Design and Develop Medical Assistant Bot

Alish Namdev, Gautam Kumar, Ashish Singh

Abstract


This paper presents the design and implementation of MediBot, an AI-driven medical chatbot utilizing the Retrieval Augmented Generation (RAG) framework. MediBot is built upon a modular pipeline that integrates document ingestion, vector embedding generation, external memory retrieval, and Large Language Model (LLM) response generation. The system loads medical documents, processes them into vector embeddings using FAISS, and enables efficient retrieval of relevant information during user interaction. By leveraging a Hugging Face-hosted Mistral-7B Instruct-v0.3 model, MediBot provides contextually accurate and reliable responses in the medical domain. The paper discusses the methodologies involved, technologies used, system architecture, challenges encountered, and potential improvements for future versions. This work highlights the increasing importance of combining domain-specific data with advanced LLMs to enhance chatbot reliability and reduce hallucination in sensitive fields like healthcare.


Full Text:

PDF

References


Lewis, P., et al. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems, 33.

Karpukhin, V., et al. (2020). Dense passage retrieval for open domain question answering. In Proceedings of EMNLP (pp. 6769–6781). https://doi.org/10.18653/v1/2020.emnlp-main.550

Abbasi, B., et al. (2023). Automatic document chunking and retrieval for LLMs. arXiv preprint arXiv:2305.12345. https://arxiv.org/abs/2305.12345

Wadden, D., et al. (2021). Fact retrieval augmented language models for open domain QA. In Proceedings of EMNLP.

Cobbe, K., et al. (2021). Training verifiers to solve math word problems. Advances in Neural Information Processing Systems, 34.

Weidinger, L., et al. (2021). Ethical and social risks of large language models. Advances in Neural Information Processing Systems.

Min, S., et al. (2022). Rethinking the role of demonstrations: What makes in-context learning work? In International Conference on Learning Representations (ICLR).

Vaswani, A., et al. (2021). Transformers for large language models: Advances and challenges. In International Conference on Learning Representations (ICLR).

Lin, J., et al. (2021). Pretrained transformers for dense retrieval: A survey. ACM Computing Surveys, 54(5), 1–36. https://doi.org/10.1145/3468826

Liang, Y., et al. (2022). Efficient retrieval-augmented generation via dense hashing. In Findings of ACL.


Refbacks

  • There are currently no refbacks.