Scholarlink

A comprehensive understanding of business processes is essential for their digitization and optimization. This thesis presents a novel approach for automatically generating Business Process Model and Notation (BPMN) 2.0 diagrams from natural language descriptions, specifically focusing on Chinese business process documents. By integrating Large Language Models (LLMs) with enhanced rule-based natural language processing (NLP) techniques, we address the complexities inherent in Chinese texts, including filtering irrelevant information, recognizing conditional sentences, and converting implicit actions into explicit ones. Our method significantly improves the accuracy and reliability of BPMN diagram generation.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

An Automatic Generation Method for Business Process Specification Based on Large Language Models

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Kai Wang,
  • Shan Li,
  • Lizong Zhang,
  • Yongjian Zhang,
  • Baobing Xia,
  • Lei Zhang,
  • Yihong Qian

摘要

A comprehensive understanding of business processes is essential for their digitization and optimization. This thesis presents a novel approach for automatically generating Business Process Model and Notation (BPMN) 2.0 diagrams from natural language descriptions, specifically focusing on Chinese business process documents. By integrating Large Language Models (LLMs) with enhanced rule-based natural language processing (NLP) techniques, we address the complexities inherent in Chinese texts, including filtering irrelevant information, recognizing conditional sentences, and converting implicit actions into explicit ones. Our method significantly improves the accuracy and reliability of BPMN diagram generation.