Developing NLP Capabilities for Quality and Risk Attribute Analysis of Requirements Specifications and Technical Documentation

开发需求规范和技术文档的质量和风险属性分析的 NLP 能力

• 批准号: 490728-2015
• 负责人: Keselj, Vlado
• 金额: $ 1.66万
• 依托单位: Dalhousie University
• 依托单位国家: 加拿大
• 项目类别: Engage Grants Program
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=586595
• 关键词: Developing; NLP; Capabilities; Quality; Risk

This project is aimed at research in the problem of assessing the quality and risk attributes relating to requirement specifications and technical documents, and providing means to trace evidence for this assessment.This analysis is of particular importance in safety-critical industries, such as aerospace, automotive, medical technology, power generation, and defence systems, where the costs of errors found throughout the development process are skyrocketing and failures after deployment can be catastrophic. The results of the research will be incorporated in improvement of the products offered by the QRA corporation, a Halifax based company, which is developing a software tool for the analysis and verification of system requirements and technical documentation written in natural language. The proposed approach exploits natural language processing techniques at the level of text classification and similarity measurements, and domain specific term extraction and semantic reasoning. The results of this research will enable the company to expand the features of the products offering to its clients, and enhancing its competitive position.

本项目旨在研究评估与以下方面有关的质量和风险属性的问题： 要求规范和技术文件，并提供追踪证据的手段 这种分析在安全关键行业中特别重要，例如航空航天，汽车， 医疗技术，发电和防御系统，其中错误的成本在整个 开发过程飞速发展，部署后的失败可能是灾难性的。的结果 研究将被纳入由QRA公司，一个哈利法克斯为基础的产品的改进 该公司正在开发一种分析和核实系统要求的软件工具， 用自然语言编写的技术文档。所提出的方法利用自然语言 在文本分类和相似性度量以及领域特定术语水平上的处理技术 提取和语义推理。这项研究的结果将使该公司能够扩大功能， 为客户提供的产品，并提高其竞争地位。