Research on Theoretical Framework for Large Model Data Governance and its constituent elements: Perspective from Standardization and Analysis of Representative Literature
摘要: 【目的/意义】旨在构建一个标准化视角下,覆盖多维度和多方面要素的大模型数据治理的理论框架,以填补当前缺少大模型数据治理专门性研究和标准化研究的空白,丰富大模型数据治理的理论研究内容,并为实际应用提供参考。【方法/过程】综合采用内容分析法和专家咨询法,对国内外相关标准、中英文代表性期刊文献进行系统梳理和分析,迭代优化并构建了大模型数据治理理论框架。【结果/结论】大模型数据治理理论框架涵盖多个维度,包括大模型数据质量管理、大模型数据管理、大模型数据资源管理、大模型数据资产管理、大模型数据风险管理等。五个维度沿着“基础要素-核心要素(执行方式、实现路径、核心目标)-保障要素”构成理论框架。
Abstract: Abstract: Purpose/Significance: This study aims to build a theoretical framework for large model data governance from a standardized perspective, covering multidimensional and multifaceted elements. This framework seeks to fill in current gaps in specialized research and standardization research on large model data governance, which will enrich theoretical research on large model data governance and provide implications to practices. Methods/Process: By holistic approach to content analysis and expert consultation, this study systematically reviews and analyzes relevant standards and representative Chinese and English journal literature, iteratively refining and constructing a theoretical framework for large model data governance. Results/Conclusions: The theoretical framework for large model data governance covers multiple dimensions, including large model data quality management, large model data management, large model data resource management, large model data asset management, and large model data risk management. This framework is constructed in alignment with "basic elements - core elements (execution methods, implementation paths, core objectives) - safeguard elements."
[V1] | 2024-08-22 14:18:52 | PSSXiv:202408.01313V1 | 下载全文 |
1. 功能重构与价值重塑:学术期刊驱动中国自主评价体系构建研究 | 2025-05-28 |
2. AIGC赋能的提示素养:生成式AI时代的人智交互能力重构 | 2025-05-26 |
3. I-model理论视角下人工智能驱动的 信息资源管理研究 | 2025-05-26 |
4. 生成式人工智能的信息资源管理职能分析——成因、挑战与应对 | 2025-05-26 |
5. 以人民为中心的档案工作底层逻辑及其实践转化 ——兼论“十五五”档案公共服务能力的提升路径 | 2025-05-21 |