面向工业互联网数据分析的机器学习工作流推荐方法  

Machine learning workflow recommendation for data analysis in industrial Internet

作　　者：文一凭[1] 田沐阳 谭铮 康国胜[1] 刘建勋[1] WEN Yiping;TIAN Muyang;TAN Zheng;KANG Guosheng;LIU Jianxun(Hunan Provincial Key Laboratory for Service computing and Novel Software Technology,Hunan University of Science and Technology,Xiangtan 411201,China;China Railway Construction Heavy Industry Corporation Limited,Changsha 410000,China)

机构地区：[1]湖南科技大学服务计算与软件服务新技术湖南省重点实验室,湖南湘潭411201 [2]中国铁建重工集团股份有限公司,长沙410000

出　　处：《计算机集成制造系统》2024年第8期2681-2687,共7页Computer Integrated Manufacturing Systems

基　　金：国家重点研发计划资助项目(2020YFB1707600);国家自然科学基金资助项目(62177014);湖南省教育厅资助项目(20B222)。

摘　　要：工业大数据具有多模态和强关联等特性,这给数据分析与应用带来了新的挑战。如何根据工业应用需求的特点实施有效的数据分析过程通常是一项非常复杂、耗时耗力的任务。针对该问题,提出一种面向工业互联网数据分析的机器学习工作流推荐方法。该方法以已有解决方案为起点,将其所使用的数据集和机器学习工作流作为推荐参考,基于Doc2vec模型与最大平均差异方法计算文本描述相似度与数据分布特征相似度,可根据当前数据分析任务需求,推荐合适的已有解决方案中的机器学习工作流。仿真实验说明了该方法的有效性。The characteristics such as multi-modality and strong association of Industrial big data have brought many challenges.How to effectively accomplish the data analysis process according to the requirements of industrial applications is a complex,time-consuming and labor-intensive task.In view of this task,a method of machine learning workflow recommendation for data analysis was proposed in industrial Internet.It started from existing solutions and utilized their involved datasets and machine learning workflows to provide recommendation.Based on Doc2vec model and the maximum average difference method,the similarities between existing solutions and the data analysis requirements by their text descriptions and data distribution features were calculated,by which suitable machine learning workflows in existing solutions could be selected and recommended.The result of simulation experiments showed effectiveness of the proposed method.

关 键 词：工业互联网 机器学习工作流 推荐 数据分析 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]