基于微调Qwen2自动构建领域UML模型

doi:10.11896/jsjkx.240900155

计算机科学 ›› 2025, Vol. 52 ›› Issue (6A): 240900155-4.doi: 10.11896/jsjkx.240900155

• 大语言模型技术及应用 • 上一篇下一篇

基于微调Qwen2自动构建领域UML模型

李嘉威¹, 邓媛丹^2,3, 陈波¹

1 电子科技大学信息与软件工程学院成都 610000
2 宜宾电子科技大学研究院四川宜宾 644000
3 厅市共建智能终端四川省重点实验室四川宜宾 644000

出版日期:2025-06-16 发布日期:2025-06-12
通讯作者: 陈波(bochen@uestc.edu.cn)
作者简介:(202322090715@std.uestc.edu.cn)
基金资助:
四川省科技计划项目(2023YFG0032);四川省科技成果转移转化示范项目(2024ZHCG0029)

Domain UML Model Automatic Construction Based on Fine-tuning Qwen2

LI Jiawei¹ , DENG Yuandan^2,3, CHEN Bo¹

1 School of Information and Software Engineering,University of Electronic Science and Technology of China,Chengdu 610000,China
2 University of Electronic Science and Technology of China Yibin Park,Yibin,Sichuan 644000,China
3 Sichuan Provincial Key Laboratory of Intelligent Terminal Jointly Built by Hall and City,Yibin,Sichuan 644000,China

Online:2025-06-16 Published:2025-06-12
About author:LI Jiawei,born in 1997,postgraduate. His main research interests include computer vision and so on.
CHEN Bo,born in 1977,Ph.D,professor. His main research interests include computational intelligence,cloud computing and intensive data processing.
Supported by:
Sichuan Province Sci-Tech Plan Project(2023YFG0032)and Sichuan Province Scientific and Technological Achievements Transfer and Transformation Demonstration Project(2024ZHCG0029).

摘要/Abstract

摘要： 提出了一种基于大模型微调技术的领域UML(统一建模语言)自动构建系统,用于将各领域软件系统制作需求的自然语言描述自动转换为符合统一建模语言标准的UML类图。研究过程包括自然文本数据集构建、模型微调、量化部署以及前端交互界面的开发。通过此系统,非专业用户可以通过简单的自然语言输入,自动生成符合统一建模语言标准的UML类图,大大降低了时间和人力成本。

关键词: 大模型微调, 领域建模, 数据集构建, 统一建模语言

Abstract: This paper proposes a domain UML(unified modeling language) automatic construction system based on large model fine-tuning technology,which is used to automatically convert natural language descriptions of software system production requirements in various domains into UML class diagrams that comply with the unified modeling language standards. The research process includes the construction of natural text datasets,model fine-tuning,quantitative deployment,and the development of front-end interactive interfaces. By this system,non-professional users can automatically generate UML class diagrams that comply with the unified modeling language standards through simple natural language input,greatly reducing time and labor costs.

Key words: Fine tuning of large models, Domain modeling, Dataset construction, Unified modeling language

中图分类号:

TP391

李嘉威, 邓媛丹, 陈波. 基于微调Qwen2自动构建领域UML模型[J]. 计算机科学, 2025, 52(6A): 240900155-4. https://doi.org/10.11896/jsjkx.240900155

LI Jiawei , DENG Yuandan, CHEN Bo. Domain UML Model Automatic Construction Based on Fine-tuning Qwen2[J]. Computer Science, 2025, 52(6A): 240900155-4. https://doi.org/10.11896/jsjkx.240900155

参考文献

[1]OUYANG L B,GUO H L. Automatic analysis modeling methodbased on structural description of domain requirements[J]. Computer Engineering and Applications,2016,52(20):52-57.
[2]ALAMI N,ARMAN N,KHAMYSEH F. A semi-automatedapproach for generating sequence diagrams from Arabic user requirements using a natural language processing tool[C]//2017 8th International Conference on Information Technology(ICIT). IEEE,2017:309-314.
[3]XU Q. The Design and Implementation of UML Diagram Construction and Content Management System Based on PlantUML[D].Nanjing:Nanjing Univercity,2020.
[4]IBRAHIM M,AHMAD R. Class diagram extraction from tex-tual requirements using natural language processing(NLP) techniques[C]//2010 Second International Conference on Computer Research and Development.IEEE,2010:200-204.
[5]WANG X. Design and Implementation of UML Model Generation Tool Based on NLP Technology[D].Beijing:Beijing University Of Posts And Teleecommunications,2020.
[6]PlantUML Language Reference Guide.Drawing UML with PlantUML[EB/OL].(2023-11)[2024-09-01].https://plantuml.com/en/guide.
[7]HAN D. Introducing Unsloth:30x faster LLM training [EB/OL].(2023-11-30)[2024-09-01].https://unsloth.ai/introducing.
[8]OMG. About the Unified Modeling Language Specification Version2.5.1[EB/OL].(2017-12)[2024-09-01].https://www.omg.org/spec/UML.
[9]GOUTTE C,GAUSSIER E.A probabilistic interpretationof precision,recall and F-score,with implication for evaluation[C]//European Conference on Information Retrieval. Berlin:Springer,2005:345-359.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

基于微调Qwen2自动构建领域UML模型

Domain UML Model Automatic Construction Based on Fine-tuning Qwen2

PDF (PC)

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

Metrics

本文评价

推荐阅读 0