系统仿真学报 ›› 2025, Vol. 37 ›› Issue (4): 943-952.doi: 10.16182/j.issn1004731x.joss.23-1435

• 论文 • 上一篇    下一篇

面向战略运筹分析的事件本体及数据集构建方法

陈泉林, 贾珺   

  1. 军事科学院 战争研究院,北京 100091
  • 收稿日期:2023-11-27 修回日期:2024-01-06 出版日期:2025-04-17 发布日期:2025-04-16
  • 通讯作者: 贾珺
  • 第一作者简介:陈泉林(1998-),男,回族,硕士生,研究方向为运筹分析。
  • 基金资助:
    国家自然科学基金(62003366);全军军事类研究生资助课题(JY2022C251)

An Event Ontology and Dataset Construction Method for Strategic Operations Analysis

Chen Quanlin, Jia Jun   

  1. Institute of War Studies, Academy of Military Sciences, Beijing 100091, China
  • Received:2023-11-27 Revised:2024-01-06 Online:2025-04-17 Published:2025-04-16
  • Contact: Jia Jun

摘要:

针对战略运筹分析领域缺少信息抽取技术研究的专业数据集的问题,提出一种面向战略运筹分析的事件本体及数据集构建方法。根据战略运筹分析中情况判断方面的需求,提出面向战略运筹分析事件本体模型,并使用“少量人工标注+微调大语言模型标注”的方法构建面向战略运筹分析的事件数据集EfSOA。数据集构建方法以及EfSOA突出了战略运筹分析领域知识,能有效支持信息抽取方法在该领域的研究,为未来构建面向战略运筹分析的事件抽取与关系挖掘模型提供方法。

关键词: 战略运筹分析, 大语言模型, 事件本体, 数据集, 信息抽取

Abstract:

Aiming at the lack of professional datasets for information extraction technology research in the field of strategic operations research analysis, this paper proposes an event ontology and dataset construction method for strategic operations research analysis. The method proposes an event ontology model for strategic operations research analysis according to the needs of situation judgment in strategic operations research analysis, and uses the method of "a small amount of manual annotation + fine-tuned large language model annotation" to construct the event dataset EfSOA for strategic operations research analysis. The dataset construction method proposed in this paper and the constructed dataset EfSOA highlight the domain knowledge of strategic operations research analysis, which can effectively support the research of information extraction methods in this field, and lay a foundation for the future construction of event extraction and relationship mining models for strategic operations research analysis.

Key words: strategic operations analysis, large language modeling, event ontology, datasets, information extraction

中图分类号: