Automated Discovery of Content-in-Context Relationships from a Large Corpus of Arctic Social Science Data

从大量北极社会科学数据中自动发现内容上下文关系

• 批准号: 1719540
• 负责人: Peter Pulsifer
• 金额: $ 29.99万
• 依托单位: University of Colorado at Boulder
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-08-01 至 2021-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1719540&HistoricalAwards=false
• 关键词: Automated; Discovery; Content; Context; Relationships

This award is for an Early Concept Grant for Exploratory Research (EAGER) project research a potentially transformative approach to information management and knowledge discovery. The research will develop an approach to searching social science data (handwritten notes, analog and digital video, photos, and audio records, maps, illustrations, etc.) without associated metadata. The approach will be tested on a unique document collection generated from the Exchange for Local Observations and Knowledge of the Arctic (ELOKA), originally funded by the National Science Foundation during the International Polar Year 2007-2008. Leveraging inherent structures and boundary conditions among embedded levels of granularity in a digital resource collection (i.e., automated granularity), this project on Automated Discovery of Content-in-Context Relationships from a Large Corpus of Arctic Social Science Data has three objectives: (1) manage a document collection without metadata or databases: (2) generate relational schema without metadata or databases; and (3) analyze efficiencies and functionalities of automated granularity relative to metadata and database solutions. Files in Portable Document Format (PDF), which is a classic type of unstructured data, will be used to generate content-in-context relationships (i.e, relational schema) objectively and simply from the inherent structure of the documents.

该奖项是为了表彰探索性研究项目研究的早期概念资助，这是一种潜在的信息管理和知识发现的变革性方法。这项研究将开发一种搜索社会科学数据(手写笔记、模拟和数字视频、照片和音频记录、地图、插图等)的方法。而没有相关联的元数据。这一方法将在北极地方观测和知识交流(EROKA)产生的独特文件集上进行测试，该交换中心最初由国家科学基金会在2007-2008年国际极地年期间提供资金。利用数字资源集合中嵌入的粒度级别之间的内在结构和边界条件(即自动粒度)，该项目从北极社会科学数据的大型语料库中自动发现上下文中的内容关系，有三个目标：(1)管理没有元数据或数据库的文档集合：(2)生成没有元数据或数据库的关系模式；以及(3)分析自动化粒度相对于元数据和数据库解决方案的效率和功能。可移植文档格式(PDF)的文件是一种经典的非结构化数据类型，它将被用来客观和简单地从文档的固有结构生成上下文中的内容关系(即关系模式)。