Evaluation and improvement of search tools for critical information

关键信息搜索工具的评估和改进

• 批准号: 5485-2011
• 负责人: Cormack, Gordon
• 金额: $ 1.75万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2013
• 资助国家: 加拿大
• 起止时间: 2013-01-01 至 2014-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=520801
• 关键词: Evaluation; improvement; search; tools; critical

We investigate the effectiveness of using search technologies to identify, as nearly as possible, all relevant documents within a very large electronic dataset, while at the same time identifying no irrelevant documents. Applications include spam filtering, legal search, medical informatics, law enforcement, security and intelligence. For these applications it is necessary not only to identify many relevant documents and few irrelevant ones, but to estimate how many of each are identified, and how many are missed.

我们调查使用搜索技术在一个非常大的电子数据集中尽可能地识别所有相关文件，同时又不识别无关文件的有效性。应用包括垃圾邮件过滤、法律搜索、医疗信息学、执法、安全和情报。对于这些应用程序，不仅需要识别许多相关的文档和少数不相关的文档，而且需要估计每个文档中有多少被识别，有多少被遗漏。