政府开放数据的数据使用状态预测模型研究

发布时间：2018-03-09 20:36

本文选题：政府开放数据　切入点：数据使用　出处：《华中师范大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着我国政府开放数据平台数量日益增多,如何有效提升政府开放数据服务水平已成为研究的热点问题。公众获取政府开放数据并进行利用是数据资源影响社会与经济发展的重要方式之一。因此,对政府开放数据的数据使用状态进行描述与预测具有现实意义。运用科学的方法合理预测政府开放数据后数据的使用情况,有利于把握公众的数据需求以及需求的变化情况,有利于提升政府数据开放平台对数据资源的整合能力,有利于提高政府开放数据的综合服务水平。本文旨在通过文献法、定性与定量相结合分析法以及实证分析法等研究方法,在分析与研究政府目前已开放的数据资源使用情况的基础上,构建政府开放数据的数据使用状态预测模型,并验证其可行性与有效性。主要研究内容分为以下三个部分:首先,提出了表示数据使用状态的两个维度。通过研究国内外相关文献,综合相关研究的特点与不足,从数据用户使用行为的角度提出政府开放数据的公众关注程度与公众利用程度两个维度,以此来表示数据资源在开放之后的使用状态。其次,构建了数据使用状态的预测模型。本文从数据属性与数据管理两个层面着手分析政府开放数据的特征,提出了使用数据领域、数据格式、数据来源、主题分类、开放条件以及更新频率等六项指标作为变量,对数据的使用状态进行预测。通过综合分析常用预测方法的优缺点,利用遗传算法优化BP神经网络模型的方法,构建政府开放数据的数据使用状态预测模型。最后,对构建的预测模型进行了实证分析。本文以上海市政府数据服务网的数据作为样本,通过实证分析对本文提出的数据使用状态表示方法以及预测模型进行了验证。实证结果表明,本文构建的政府开放数据的数据使用状态预测模型具备一定的可行性,同时从定量的角度客观的反映了上述方法与模型的有效性。本文的研究为政府更好把握政府数据资源的使用情况提供了可行的预测分析方法,有助于提升政府开放数据综合服务水平,同时也为今后在政府开放数据预测方面的研究提供理论与实践依据。
[Abstract]:With the increasing number of open data platforms in our government, How to effectively improve the level of government open data services has become a hot issue. Public access to open government data and use of data resources is one of the important ways to affect social and economic development. It is of practical significance to describe and predict the state of data use of government open data. The use of scientific methods to predict the use of data after the opening of government data is beneficial to grasp the data demand of the public and the changing situation of demand. This paper aims to improve the integration ability of the open platform of government data to the data resources, and to improve the comprehensive service level of the open data of the government. This paper aims at the research methods of literature method, qualitative and quantitative analysis method and empirical analysis method, etc. On the basis of analyzing and studying the utilization of the data resources that the government has already opened, this paper constructs the prediction model of the data usage status of the government open data, and verifies its feasibility and validity. The main research contents are as follows: first, Two dimensions are proposed to represent the state of data use. Through the study of relevant literature at home and abroad, the characteristics and shortcomings of related studies are summarized. From the point of view of data users' use behavior, this paper puts forward two dimensions of public concern and public utilization of government open data, so as to express the status of data resources after opening up. Secondly, This paper analyzes the characteristics of government open data from the two aspects of data attribute and data management, and proposes the use of data domain, data format, data source, subject classification. Six indexes, such as open condition and update frequency, are used as variables to predict the status of data use. By synthetically analyzing the advantages and disadvantages of common prediction methods, the genetic algorithm is used to optimize the BP neural network model. Finally, the paper makes an empirical analysis of the prediction model. This paper takes the data of Shanghai Municipal Government data Service Network as a sample. Through the empirical analysis, the paper verifies the data usage state representation method and the prediction model proposed in this paper. The empirical results show that the data use state prediction model constructed in this paper is feasible. At the same time, it objectively reflects the validity of the above methods and models from a quantitative point of view. The research in this paper provides a feasible method of prediction and analysis for the government to better grasp the use of government data resources. It is helpful to improve the comprehensive service level of government open data, and to provide theoretical and practical basis for the future research on the prediction of government open data.
【学位授予单位】：华中师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：D63

【参考文献】