摘要
从基于自然语言的需求文本中抽取概念模型已有很多相关研究,然而,抽取模型中的关系信息因其复杂性而较少被研究者系统地分析和处理.文中提出了一个通用的关系信息抽取方法,给出抽取规则,从需求文本中确定和抽取关系信息.基于该方法设计并实现了一个系统CREAT3,从中文需求文本自动生成i*框架中的SD(StrategyDependency,策略依赖)模型,侧重抽取策略依赖关系信息.将得到的模型和专家抽取结果进行对比,结果显示该系统可以获得相当高的准确率,同时也保证了很高的召回率,证明了方法的可用性.并且较相关工作具有更好的可扩展性.
Much has been written on the processing of NI.（natural language）-based requirements documents to yield conceptual models. However, the extraction of links in models is deemed difficult. In this paper, we propose an approach for identifying and extracting relations in a range of requirements documents using linguistic clues and pattern-matching. The set of linguistic pat terns used for identifying relations was based on a thorough review of the literature and on experi- ments to sample requirements corpuses. Based on this approach, we developed a system to auto- matically generate the SD model of i ~ framework from Chinese requirements documents. A series of experiments were conducted to evaluate the performance of the automated requirements analy- sis system. The results show that the system achieves high recall with a consistent improvement in precision, which demonstrates the applicability of our approach.
出处
《计算机学报》
EI
CSCD
北大核心
2013年第1期54-62,共9页
Chinese Journal of Computers
基金
国家"九七三"重点基础研究发展规划项目基金(2009CB320706)
国家"八六三"高技术研究发展计划项目基金(2012AA040904)
国家自然科学基金重大项目(90818026)资助~~
关键词
自然语言需求文本
依存文法
关系抽取
策略依赖模型
NL-based requirements
dependency grammar
relation elicitation
strategic dependency model
