Subjects: Library Science，Information Science >> Library Science submitted time 2023-10-08 Cooperative journals: 《知识管理论坛》
Abstract： [Purpose/significance] Open government data is conducive to the rational development and utilization of data resources. It can encourage social innovation and promote economic development. Besides, in order to ensure effective utilization and social increment of open government data, high-quality metadata schemes is necessary. [Method/process] Firstly, this paper analyzed the related research of open government data at home and abroad. Then, it investigated the open metadata schemes of some Chinese main local governments’ data platforms, and made a comparison with the metadata standard of American open government data. [Result/conclusion] This paper reveals that there are some disadvantages about Chinese local government open data affect the use effect of open data, which including that different governments use different data metadata schemes, the description of data set is too simple for further utilization and usually presented in HTML Web page format with lower machine-readable. Therefore, our government should come up with a standardized metadata schemes by drawing on the international mature and effective metadata standard, to ensure the social needs of high quality and high value data.
Subjects: Library Science，Information Science >> Information Science submitted time 2017-12-05 Cooperative journals: 《数据分析与知识发现》
Abstract：【目的】通过网络用户评论, 为评论网站构建有效的评分预测机制。【方法】提出基于网络用户评论的评 分预测模型, 该模型包括 4 个模块: 网络用户评论获取模块、预测变量获取模块、预测分析模块以及预测结果评 价模块。抓取 30 部不同类型的电影评论数据, 27 部用于构建模型, 3 部用于检验模型。【结果】使用逐步回归方 法筛选出变量: 参与评分人数、参与评论人数、想要观看人数和电影正向评论情感均值, 构建评分预测模型。使 用 3 部电影验证, 预测评分与 IMDb 评分相差最大值为 0.0644, 最小值为 0.0227。【局限】在数据样本量、情感 特征提取精度、模型普适性验证等方面有待进一步提升。【结论】该模型能够依据用户评论对评分进行有效预测, 在网络水军探测方面也能发挥一定的作用。
Subjects: Library Science，Information Science >> Information Science submitted time 2017-11-30 Cooperative journals: 《数据分析与知识发现》
Abstract：【目的】通过网络用户评论, 为评论网站构建有效的评分预测机制。【方法】提出基于网络用户评论的评 分预测模型, 该模型包括 4 个模块: 网络用户评论获取模块、预测变量获取模块、预测分析模块以及预测结果评 价模块。抓取 30 部不同类型的电影评论数据, 27 部用于构建模型, 3 部用于检验模型。【结果】使用逐步回归方 法筛选出变量: 参与评分人数、参与评论人数、想要观看人数和电影正向评论情感均值, 构建评分预测模型。使 用 3 部电影验证, 预测评分与 IMDb 评分相差最大值为 0.0644, 最小值为 0.0227。【局限】在数据样本量、情感 特征提取精度、模型普适性验证等方面有待进一步提升。【结论】该模型能够依据用户评论对评分进行有效预测, 在网络水军探测方面也能发挥一定的作用。
Hits
YES
Hits
Downloads
Comment