Method for Constructing Machine Learning Project Canvas Based on Enterprise Architecture Modeling
Method for Constructing Machine Learning Project Canvas Based on Enterprise Architecture Modeling
复制标题
DOI：
10.1016/j.procs.2022.09.077
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
H. Takeuchi;Yu Ito;Shuichiro Yamamoto
中科院分区：
文献类型：
--
作者：
H. Takeuchi;Yu Ito;Shuichiro Yamamoto
在这项研究中，我们考虑使用机器学习（ML）技术开发系统的项目。已经提出了一个ML项目画布来表示项目，以便利益相关者可以对项目有一个共同的理解。在许多情况下，这个画布必须由业务部门的从业人员在没有数据科学家足够支持的情况下构建，画布模型的质量取决于从业人员的技能或经验。因此，我们提出了一种使用业务-AI对齐模型构建特定于项目的项目画布模型的方法，并通过对ML项目实践的分析来确认该方法的有效性。
In this study, we consider projects in which systems are developed using machine learning (ML) techniques. An ML project canvas has been proposed to represent the project so that stakeholders can have a common understanding of the project. In many cases, this canvas must be constructed by business division practitioners without sufficient support from data scientists and the quality of the canvas model is dependent on the skills or experience of the practitioner. Therefore, we propose a method for constructing a project-specific project canvas model using the business–AI alignment model, and confirm the effectiveness of the method through the analysis of ML project practices.