科研管理

科研管理 ›› 2026, Vol. 47 ›› Issue (1): 111-120.DOI: 10.19571/j.cnki.1000-2995.2026.01.011

• 论文 • 上一篇    下一篇

在线创新社区知识共创过程研究——机器学习和系统动力学的整合视角

汪万1，蔡三发1,2,3   

  1. 1.同济大学经济与管理学院，上海200092;
    2.同济大学高等教育研究所，上海200092;
    3.同济大学政治与国际关系学院，上海200092
  • 收稿日期:2024-03-06 修回日期:2024-12-19 接受日期:2024-12-19 出版日期:2026-01-20 发布日期:2026-01-12
  • 通讯作者: 蔡三发
  • 基金资助:
    国家自然科学基金面上项目：“面向大中学智慧衔接的动态学生画像和智能学业规划”（62177036，2022.01—2025.12）。

Research on knowledge co-creation process in online innovation communities: An integrated perspective of machine learning and system dynamics

Wang Wan1, Cai Sanfa1,2,3

  1. 1.School of Economics and Management, Tongji University, Shanghai 200092, China; 
    2.Institute of Higher Education, Tongji University, Shanghai 200092, China;
    3. School of Political Science & International Relations, Tongji University, Shanghai 200092, China

  • Received:2024-03-06 Revised:2024-12-19 Accepted:2024-12-19 Online:2026-01-20 Published:2026-01-12

摘要：    数字技术赋能的在线创新社区是企业获取用户知识贡献的重要平台，而如何汇聚用户知识并与之共创出新知识尚未得到系统化的解析。因此，本研究结合机器学习和系统动力学，辅以量子力学测量相关因素，以构建知识共创系统动力学模型，进而探究在线创新社区知识共创过程机理。研究结果表明，在线创新社区知识共创是一个多过程相互依赖的综合体，包括知识共享、知识整合和知识创造，并呈现出差异性的知识活动特征。企业知识创新速度较用户知识创新速度更高，导致企业知识存量增速和总量均高于用户；知识共享频率、知识兼容系数和知识创造效率分别正向影响与之相应的知识共创过程；社区氛围分别与知识共享、知识整合和知识创造呈正相关，且对知识创造的影响最大，而知识共享受影响最小。本研究不仅解构了在线创新社区知识共创过程，拓展了知识共创的研究视角，也在实践层面为社区知识管理提供了启发。

关键词: 知识共创, 在线创新社区, 主题建模, 系统动力学

Abstract:   Digital technology-enabled online innovation community (OIC) is a crucial platform for enterprises to obtain users′ knowledge contributions. However, the systematic analysis of how to aggregate users′ knowledge and co-create new knowledge with them remains absent. Consequently, this study combined machine learning and system dynamics, supplemented with quantum mechanics to measure relevant factors, to construct a system dynamics model of knowledge co-creation and thereby explored the mechanism of OIC knowledge co-creation process. The research findings suggested that OIC knowledge co-creation is a multi-process interdependent complex, encompassing knowledge sharing, knowledge integration, and knowledge creation, and exhibiting differential knowledge activity characteristics. The rate of enterprise knowledge innovation is higher than that of user knowledge innovation, leading to a higher growth rate and total amount of enterprise knowledge stock compared to users. The frequency of knowledge sharing, the knowledge compatibility coefficient, and the knowledge creation efficiency positively affect the corresponding knowledge co-creation processes respectively. The community atmosphere is positively correlated with knowledge sharing, knowledge integration, and knowledge creation respectively, with the greatest impact on knowledge creation and the least on knowledge sharing. This study has not only decomposed the process of OIC knowledge co-creation and broadened the research perspective of knowledge co-creation but also will offer inspiration for community knowledge management at the practical level.

Key words: knowledge co-creation, online innovation community (OIC), topic modelling, system dynamics