Collaborative Research: Automated Real-time Production of Political Indicators

协作研究：自动实时生成政治指标

• 批准号: 1259266
• 负责人: Michael Ward
• 金额: $ 30.21万
• 依托单位: Duke University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-04-01 至 2017-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1259266&HistoricalAwards=false
• 关键词: Collaborative; Research; Automated; Real; time

Systematic indicators of political characteristics such as regime type, internal instability, and inter-state conflict have been widely used in political and economic research over the past three decades and are increasingly used in policy applications. There has been limited effort to automate the production of such indicators, which are generally produced by individual research projects using human coders and are widely available only after a considerable delay. Improvements in computer software and hardware, meanwhile, have made the process of analyzing large bodies of text to extract meaningful information much more efficient. Automated coding is transparent, without the unreproducible subjective elements of human coding, and once the source texts have been prepared, recoding to account for new theoretical or technological perspectives can be done quickly and efficiently. Using extant tools for automated coding, this project will produce near real-time political indicators that will be comparable, if not identical, to indicators produced by human coding. The project first creates data on the interaction events of nations in the current century. In addition, the project will extract measurements of the characteristics and behaviors of governments, including human rights practices and the level of autocracy. Finally, the project will construct information about the interstate disputes among nations.In the realm of world politics, most scientific analysis is based on data that are at least a decade old. This research will significantly reduce the time-lag between events in the contemporary world and systematic measurement and analysis of those events in the academic and policy arena by reducing to weeks from decades the delay with which systematic data are made publicly available. Moreover, this project will expose these data in a transparent and widely accessible way. Not only will the data be available to those who collected the data, but they will be freely accessible to any and all via a web interface. The project will permit recent patterns and changes, such as the rise of China, to be more readily reflected in scientific research on world affairs. This should provide a continuing foundation for more accurate and increasingly credible understandings of politics at the global level.

在过去的三十年里，政治特征的系统指标，如政权类型，内部不稳定和国家间冲突，已被广泛用于政治和经济研究，并越来越多地用于政策应用。 这些指标通常是由个别研究项目使用人工编码人员编制的，经过相当长的拖延后才广泛提供，因此在自动编制这些指标方面所作的努力有限。 与此同时，计算机软件和硬件的改进使得分析大量文本以提取有意义信息的过程更加有效。 自动化编码是透明的，没有人类编码不可复制的主观因素，一旦源文本已经准备好，重新编码以解释新的理论或技术观点可以快速有效地完成。 利用现有的自动编码工具，该项目将产生接近实时的政治指标，这些指标即使不完全相同，也将与人工编码产生的指标相当。 该项目首先创建有关本世纪各国互动事件的数据。 此外，该项目还将提取政府特征和行为的衡量标准，包括人权实践和专制程度。最后，该项目将构建有关国家间争端的信息。在世界政治领域，大多数科学分析都是基于至少十年前的数据。 这项研究将大大缩短当代世界的事件与学术和政策竞技场对这些事件进行系统测量和分析之间的时间差，将系统数据公布的延迟从几十年缩短到几周。 此外，该项目还将以透明和可广泛获取的方式公布这些数据。 这些数据不仅将提供给收集数据的人，而且任何人都可以通过网络界面免费访问。 该项目将使最近的模式和变化，如中国的崛起，更容易反映在对世界事务的科学研究中。 这应该为更准确和更可信地理解全球一级的政治提供一个持续的基础。