Workshop: Tools for Text: Annotation Methods for Political Science Research

研讨会：文本工具：政治科学研究的注释方法

• 批准号: 0926147
• 负责人: John Wilkerson
• 金额: $ 4.93万
• 依托单位: University of Washington
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2009
• 资助国家: 美国
• 起止时间: 2009-09-01 至 2010-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0926147&HistoricalAwards=false
• 关键词: Workshop; Tools; Text; Annotation; Methods

Digitization is dramatically altering research demands and opportunities in political science. Governments around the world now release large volumes of digitized text on a daily basis, while national projects in many countries are now digitally scanning centuries of historical government documents. Online media developments are also dramatically increasing the amount and diversity of accessible digital political content, from the advent of 24 hour news coverage in mainstream media outlets, to campaign websites, political blogs, social networks and even YouTube.These developments are creating unprecedented research opportunities for political scientists. However, a central constraint to the study of text is the need to convert textual data to useful information about political patterns. Ultimately, scholars are interested in text for what it can tell us about the topics government addresses, relations among government actors, or which frames seems to carry the most weight, to name just a few examples. To answer these types of questions, approaches to categorizing text must be developed and applied.This workshop will equip newer faculty and advanced graduate students with the knowledge and skills required to incorporate well-established text annotation methods into their teaching and research. The workshop will be held at the University of Washington in Seattle and will include several introductory presentations by experts in different methods, including intensive focus on training and application of these tools. Participants will leave with an understanding the range of options available, the strengths and limits of different approaches, and actual experience in applying and evaluating them.

数字化正在极大地改变政治学的研究需求和机会。世界各地的政府现在每天都会发布大量的数字化文本，而许多国家的国家项目现在正在对数百年的历史政府文件进行数字扫描。在线媒体的发展也极大地增加了可访问的数字政治内容的数量和多样性，从主流媒体24小时新闻报道的出现，到竞选网站、政治博客、社交网络甚至YouTube，这些发展为政治科学家创造了前所未有的研究机会。然而，文本研究的一个核心限制是需要将文本数据转换为有关政治模式的有用信息。最终，学者们对文本感兴趣，因为它可以告诉我们关于政府地址的主题，政府行为者之间的关系，或者哪些框架似乎最有分量，仅举几例。为了回答这些类型的问题，必须开发和应用文本分类的方法。本研讨会将为新教师和高级研究生提供将成熟的文本注释方法纳入教学和研究所需的知识和技能。讲习班将在西雅图的华盛顿大学举行，专家将以不同方法作介绍性发言，包括集中讨论这些工具的培训和应用。与会者将了解现有的选择范围，不同方法的优势和局限性，以及应用和评估它们的实际经验。