高级检索
    裴忠一, 刘璘, 王晨, 王建民. 面向机器学习应用的可解释性需求分析框架[J]. 计算机研究与发展. DOI: 10.7544/issn1000-1239.202220794
    引用本文: 裴忠一, 刘璘, 王晨, 王建民. 面向机器学习应用的可解释性需求分析框架[J]. 计算机研究与发展. DOI: 10.7544/issn1000-1239.202220794
    Pei Zhongyi, Liu Lin, Wang Chen, Wang Jianmin. An Explainability-Centric Requirements Analysis Framework for Machine Learning Applications[J]. Journal of Computer Research and Development. DOI: 10.7544/issn1000-1239.202220794
    Citation: Pei Zhongyi, Liu Lin, Wang Chen, Wang Jianmin. An Explainability-Centric Requirements Analysis Framework for Machine Learning Applications[J]. Journal of Computer Research and Development. DOI: 10.7544/issn1000-1239.202220794

    面向机器学习应用的可解释性需求分析框架

    An Explainability-Centric Requirements Analysis Framework for Machine Learning Applications

    • 摘要: 基于大数据机器学习的智能软件研发过程需要综合运用软件工程、数据与领域知识工程、机器学习等多方面的知识和工具,涉及的研究主题和人员角色众多,技术实现手段复杂,研发难度大. 面向智能软件的需求工程因此需要面对领域知识、业务知识、数据科学交织带来的挑战. 然而,如何将领域知识和端到端的机器学习技术恰当地融合到给定的业务流程之中,以及如何应对工业、医疗等高可信要求场景中的可解释性需求,仍是亟待探索的重要研究问题. 调研了近年来面向机器学习应用的需求工程研究文献,对该领域的发展现状、核心问题和代表性方法进行综述. 归纳后,提出了面向机器学习应用的可解释性需求分析框架. 基于该框架,通过一个工业智能应用案例分析了未来待研究的重要问题,展望了可行的研究路径.

       

      Abstract: Data-driven intelligent software based on machine learning technology is an important means to realize industrial digital transformation. Its research and development processes require the combined use of software requirements engineering, data and domain knowledge engineering, machine learning and so on. This process involves many subjects and roles, making it extremely challenging to clearly explain why and how the domain knowledge, business logic and data semantics relate to each other. Hence, a systematic requirements engineering approach is needed to explicitly address the explainability requirements issues of data-driven intelligence applications. It is still a fast-evolving research field which requires the proper embedding of various domain models and end-to-end machine learning technology into a given business processes. A key research question is how to deal with explainability as a core requirement for safety-critical scenarios in industrial, medical and other applications. This paper provides a research overview on requirements engineering for machine learning applications, in relation to explainability. First, the research status quo, research foci and representative research progress are reviewed. Then, an explainability-centric requirements engineering framework for machine learning applications is proposed, and some open important issues are put forward. Finally, based on the proposed framework, a case study of industrial intelligence application is discussed to illustrate the proposed requirements analysis methodological framework.

       

    /

    返回文章
    返回