Back to List
Notice:This resource is provided by a third-party author. Please review the code with AI tools or manually before use to ensure security and compatibility.
Jupyter Notebookhoneyandme/RAGQnASystem

RAGQnASystem

本项目设计并实现了一个基于知识图谱 RAG 与大语言模型的医疗智能问答系统。系统以 DiseaseKG 医疗数据集为基础,依托 Neo4j 构建包含 4.4 万实体与 31 万关系的领域知识图谱;结合 BERT 命名实体识别与 32B 大模型意图识别,通过精确的图谱检索与受控的答案生成,有效缓解大模型在医疗场景中的幻觉问题,显著提升回答的准确性与可靠性。

51.6/100
1.3KForks: 140
View on GitHub
Loading report...

Similar Projects

generative-ai-for-beginners

80

21 Lessons, Get Started Building with Generative AI

Jupyter Notebook111.8K

LLMs-from-scratch

81

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Jupyter Notebook96.9K

openai-cookbook

83

Examples and guides for using the OpenAI API

Jupyter Notebook74.1K

ai-agents-for-beginners

77

12 Lessons to Get Started Building AI Agents

Jupyter Notebook66.8K
Back to List