CAREER: An Integrated Framework for Extracting and Utilizing Information from Tables and Figures in Digital Libraries

职业：从数字图书馆的表格和图形中提取和利用信息的集成框架

• 批准号: 0845487
• 负责人: Prasenjit Mitra
• 金额: $ 44.98万
• 依托单位: Pennsylvania State Univ University Park
• 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: 2009-06-01 至 2014-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0845487&HistoricalAwards=false
• 关键词: CAREER; Integrated; Framework; Extracting; Utilizing

"This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)."This project is concerned with retrieval of information contained in the tables, figures and references contained in digital document on the web. At present a large amount of information present in tables and figures in documents is not easy accessible by automated methods. End-users cannot confidently pose a query and retrieve data presented in tables and figures in documents directly through a search engine. While there has been substantial research on table boundary detection and some research on table content extraction, none of these methods tried to identify the data and segregate it from the metadata present in the tables and figures which would represent a significant step forward. The research goal of this proposal is to make fundamental contributions in digital document processing and information retrieval. The objective is to identify data clearly, separate it from the metadata automatically, and store in a digital repository. An end-user will then be able to effectively query and retrieve datasets or their interesting parts.

“这项奖励是根据2009年美国复苏和再投资法案（公法111-5）资助的。”这个项目是关于检索包含在网络数字文档中的表格、图表和参考文献中的信息。目前，文件中表格和图表中的大量信息不容易通过自动化方法获取。最终用户不能自信地提出查询，并直接通过搜索引擎检索文档中表格和图形中呈现的数据。虽然对表边界检测和表内容提取进行了大量的研究，但这些方法都没有尝试识别数据并将其从表和图中的元数据中分离出来，这将是向前迈出的重要一步。本课题的研究目标是在数字文档处理和信息检索方面做出基础性的贡献。目标是清晰地识别数据，自动将其与元数据分离，并存储在数字存储库中。最终用户将能够有效地查询和检索数据集或其感兴趣的部分。