基于大模型检索增强生成的气象数据库问答模型实现

doi:10.3778/j.issn.1002-8331.2406-0230

摘要/Abstract

摘要： 随着信息检索和知识获取需求的增加，智能问答系统在多个垂直领域得到广泛应用。然而，在气象领域仍缺乏专门的智能问答系统研究，严重限制了气象信息的高效利用和气象系统的服务效率。针对这一需求，提出了一种面向气象数据库的大模型检索智能问答技术实现方案。该方案设计了一种基于关系型数据库（SQL）与文档型数据（NoSQL）的多通道查询路由（multi-channel retrieval router，McRR）方法，为了适配数据库进行大模型查询以及增强大模型对查询表的理解，分别提出指令查询转换方法与数据库表摘要方法DNSUM，提升大模型对数据库的语义理解能力，通过结合问题理解、重排序器和响应生成等关键模块，构建了一个端到端的智能问答模型，可实现多数据源的相关知识检索及答案生成。实验结果显示，该模型可以有效理解用户问题并生成准确的答案，具有良好的检索和响应能力。不仅为气象领域提供了一种智能问答的解决方案，也为气象智能问答技术提供了新的应用实施参考。

关键词: 数据库查询, 数据库问答, 大语言模型, 检索增强生成, 气象问答

Abstract: With the increasing demand for information retrieval and knowledge acquisition, question-answering systems are widely applied across various domains. However, there is a notable lack of specialized question-answering system research in the meteorological field, which severely limits the efficient utilization of meteorological information and the service efficiency of meteorological systems. To address this gap, it proposes a retrieval-augmented generation based question-answering implementation scheme for meteorological databases. This scheme designs a multi-channel query routing （McRR） method based on relational databases （SQL） and document-oriented data （NoSQL）. Additionally, to adapt large model queries to databases and enhance the model’s understanding of query tables, the paper proposes an instruction query conversion method and a database table summarization method （termed as DNSUM） to improve the model’s semantic understanding of databases. Furthermore, by integrating key modules such as question understanding, re-rankers, and response generation, it constructs an end-to-end intelligent question-answering engine capable of retrieving relevant knowledge and generating answers from multiple data sources. Experimental results on the constructed meteorological question-answering dataset demonstrate that this engine effectively understands user queries and generates accurate answers, exhibiting strong retrieval and response capabilities. This research not only provides a question-answering solution for the meteorological field but also offers new directions for the application of question-answering technology in vertical domains.

Key words: sructured query, database question-answering, large language models（LLM）, retrieval-augmented generation （RAG）, meteorological Q&, A

江双五, 张嘉玮, 华连生, 杨菁林. 基于大模型检索增强生成的气象数据库问答模型实现[J]. 计算机工程与应用, 2025, 61(5): 113-121.

JIANG Shuangwu, ZHANG Jiawei, HUA Liansheng, YANG Jinglin. Implementation of Meteorological Database Question-Answering Based on Large-Scale Model Retrieval-Augmentation Generation[J]. Computer Engineering and Applications, 2025, 61(5): 113-121.

参考文献

[1] BROWN T, MANN B, RYDER N, et al. Language models are few-shot learners[C]//Advances in Neural Information Processing Systems, 2020: 1877-1901.
[2] OPENAI. GPT-4 technical report[J]. arXiv:2303.08774, 2023.
[3] OPENAI. Introducing ChatGPT [EB/OL].(2022-11-30)[2024-04-17]. https: //openai.com/blog/chatgpt.
[4] OUYANG L, WU J, JIANG X, et al. Training language models to follow instructions with human feedback[C]//Advances in Neural Information Processing Systems, 2022: 27730-27744.
[5] TOUVRON H, LAVRIL T, IZACARD G, et al. LLaMA: open and efficient foundation language models[J].arXiv: 2302.13971,2023.
[6] TAORI R, GULRAJANI I, ZHANG T, et al. Stanford Alpaca: an instruction-following LLaMA model [EB/OL]. (2023-03-13)[2024-04-13]. https://crfm.stanford.edu/2023/03/13/alpaca.html.
[7] JIN B, ZENG H, WANG G, et al. Language models as semantic indexers[C]//Proceedings of the 42nd International Conference on Machine Learning, 2024: 1-16.
[8] 车万翔, 窦志成, 冯岩松, 等. 大模型时代的自然语言处理: 挑战、机遇与发展[J]. 中国科学: 信息科学, 2023,53 (9): 1645-1687.
CHE W X, DOU Z C, FENG Y S, et al. Towards a comprehensive understanding of the impact of large language models on natural language processing: challenges, opportunities and future directions[J]. SCIENTIA SINICA Informationis, 2023, 53(9): 1645-1687.
[9] 张鹤译, 王鑫, 韩立帆, 等. 大语言模型融合知识图谱的问答系统研究[J]. 计算机科学与探索, 2023, 17 (10): 2377-2388.
ZHANG H Y, WANG X, HAN L F, et al. Research on question answering system on joint of knowledge graph and large language models[J].Journal of Frontiers of Computer Science & Technology, 2023, 17(10): 2377-2388.
[10] LEWIS P, PEREZ E, PIKTUS A, et al. Retrieval-augmented generation for knowledge-intensive NLP tasks[C]//Advances in Neural Information Processing Systems, 2020: 9459-9474.
[11] MAO Y, HE P, LIU X, et al. Generation-augmented retrieval for open-domain question answering[J]. arXiv:2009.08553, 2020.
[12] GUU K, LEE K, TUNG Z, et al. Retrieval augmented language model pre-training[C]//Proceedings of the International Conference on Machine Learning, 2020: 3929-3938.
[13] JIANG Z, XU F F, GAO L, et al.Active retrieval augmented generation[J]. arXiv: 2305.06983, 2023.
[14] ZHAO P, ZHANG H, YU Q, et al. Retrieval-augmented generation for AI-generated content: a survey[J]. arXiv:2402.
19473, 2024.
[15] CHEN J, LIN H, HAN X, et al. Benchmarking large language models in retrieval-augmented generation[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2024: 17754-17762.
[16] UPTRAIN. AI Uptrain[EB/OL].(2021-01-01)[2023-12-09]. https://demo.uptrain.ai/benchmark.
[17] XIAO S, LIU Z, ZHANG P, et al. C-pack: packaged resources to advance general Chinese embedding[J]. arXiv:2309.07597, 2023.
[18] XIAO S, LIU Z, SHAO Y, et al. RetroMAE: pre-training retrieval-oriented language models via masked auto-encoder[J]. arXiv: 2205.12035, 2022.
[19] YANG A, XIAO B, WANG B, et al. Baichuan 2: open large-scale language models[J]. arXiv:2309.10305, 2023.
[20] Cohere. Cohere embeddings[EB/OL]. (2023-01-01)[2024-05-13]. https://cohere.com.
[21] Mistral AI. Mistral AI embeddings[EB/OL].(2023-09-27) [2024-03-13]. https://mistral.ai/news/announcing-mistral-7b.
[22] NEELAKANTAN A, XU T, PURI R, et al. Text and code embeddings by contrastive pre-training[J]. arXiv:2201.10005, 2022.
[23] RADFORD A, WU J, CHILD R, et al. Language models are unsupervised multitask learners[J]. OpenAI Blog, 2019, 1(8): 9.
[24] DU Z, QIAN Y, LIU X, et al. GLM: general language model pretraining with autoregressive blank infilling[C]//Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics, 2022: 320-335.
[25] ZENG A, LIU X, DU Z, et al. Glm-130b: an open bilingual pre-trained model[J]. arXiv: 2210.02414, 2022.
[26] DEFOG INC. SQLCoder[EB/OL].(2023-09-27)[2024-01-01]. https://defog.ai/sqlcoder-demo.
[27] CHEN J, XIAO S, ZHANG P, et al. M3-embedding: multi-linguality, multi-functionality, multi-granularity text embeddings through self-knowledge distillation[C]//Findings of the Association for Computational Linguistics, 2024: 2318-2335.