节点文献
基于ARIMA模型的政府公共数据开放预测研究
Prediction of Government Public Data Opening Based on ARIMA Model
【摘要】 【目的/意义】推进公共数据开放,已成为推进国家治理体系和治理能力现代化的重要手段,是发挥公共数据要素价值的必由之路。【方法/过程】选取上海市政府公共数据,利用差分自回归移动平均(Autoregressive Integrated Moving Average,ARIMA)模型来预测政府2024年4月22日后发布的公共数据量。【结果/结论】将模型预测结果与已有公共数据总量进行比较,得出政府公共数据中各领域排序,并将其与问卷调查得到的排序结果进行对比,得出公众心中对各领域的了解与满意程度和政府机关在这些领域上的投入不平衡。在需求预测分析、基础设施建设、政策标准制定、数据产品供给等方面为公共数据开放提出了建议,同时为政府公共数据开放提供了一定的理论和实践依据。【创新/局限】提出一种政府公共数据开放预测模型,主要局限是选取数据的最后更新时间作为数据的发布时间这一前提,并未研究数据更新这一动态过程。
【Abstract】 【Purpose/significance】Promoting the government public data opening has become an important method to modernize the governance system and governance capacity, and is also a necessary way to better utilize the value of public data as a production factor.【Method/process】Shanghai public data is selected and Autoregressive Integrated Moving Average(ARIMA) model is utilized to predict the amount of data released after April 22, 2024.【Result/conclusion】Compared the predicted results of the model with the total amount of existing public data, we obtain the ranking of various fields in government public data. The differences between the results and the ranking results obtained from the questionnaire survey reflect the imbalance between the public’s understanding and satisfaction with various fields and the government’s investment in these field. Suggestions have been put forward for public data opening in areas such as demand forecasting analysis, infrastructure construction, policy standard formulation, and data product supply, while providing theoretical and practical basis for government public data opening.【Innovation/limitation】The paper puts forward a model on prediction of government public data opening. The main limitation is that the model selects the last update time of data as the premise of data release time, and does not study the dynamic process of data update.
【Key words】 government public data; public data opening; ARIMA model; prediction; ranking;
- 【文献出处】 情报科学 ,Information Science , 编辑部邮箱 ,2025年06期
- 【分类号】G259.2;D63
- 【下载频次】66