基于本地RAG的井控知识问答系统的实现
Journal: Project Engineering DOI: 10.12238/pe.v2i5.9882
Abstract
构建智能高效的井控专业知识问答系统可有效帮助企业进行员工培训,并进行实时井控作业在线指导。基于大语言模型(large language model,LLM)的知识问答系统相较传统方法具有更高的灵活性和丰富性,检索增强生成(Retrieval-Augmented Generation,RAG)技术是一种结合外部知识库与LLM构建知识问答系统的有效方法。本研究应用了开源RAG框架Dify与大模型部署平台Ollama搭建了本地化的井控知识问答系统,实现了井控知识的精准高效问答并可保护企业的隐私数据,具有较强的应用价值。
Keywords
人工智能RAG井控知识问答系统
Full Text
PDF - Viewed/Downloaded: 0 TimesReferences
[1] 喻海霞.井控培训在井控风险控制中的重要性[J].西部探矿工程,2014,26(05):19-22.
[2] 张金营,王天堃,么长英,等.基于大语言模型的电力知识库智能问答系统构建与评价[J/OL].计算机科学,1-10[2024-07-14].https://libresource.xiyi.edu.cn:443/http/80/net/cnki/kns/yitlink/kcms/detail/50.1075.TP.20240528.0931.002.html.
[3] 齐思洋,胡慧云,李洪冰,等.融合大语言模型的领域问答系统构建方法[J/OL].北京邮电大学学报,1-7[2024-07-14]. https://libresource.xiyi.edu.cn:443/https/443/org/doi/yitlink/10.13190/j.jbupt.2023-279.
[4] Lewis P, Perez E, Piktus A, et al. Retrieval-augmented generation for knowledge-intensive nlp tasks[J].Advances in Neural Information Processing Systems,2020,33:9459-9474.
[5] Edge D,Trinh H,Cheng N,etal.From local to global:A graph rag approach to query-focused summarization[J].arxiv preprint arxiv:2404.16130,2024.
[2] 张金营,王天堃,么长英,等.基于大语言模型的电力知识库智能问答系统构建与评价[J/OL].计算机科学,1-10[2024-07-14].https://libresource.xiyi.edu.cn:443/http/80/net/cnki/kns/yitlink/kcms/detail/50.1075.TP.20240528.0931.002.html.
[3] 齐思洋,胡慧云,李洪冰,等.融合大语言模型的领域问答系统构建方法[J/OL].北京邮电大学学报,1-7[2024-07-14]. https://libresource.xiyi.edu.cn:443/https/443/org/doi/yitlink/10.13190/j.jbupt.2023-279.
[4] Lewis P, Perez E, Piktus A, et al. Retrieval-augmented generation for knowledge-intensive nlp tasks[J].Advances in Neural Information Processing Systems,2020,33:9459-9474.
[5] Edge D,Trinh H,Cheng N,etal.From local to global:A graph rag approach to query-focused summarization[J].arxiv preprint arxiv:2404.16130,2024.
Copyright © 2024 刘月月
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License