Towards Evidence-Based Discovery

海量的电子信息为科学家确定重大挑战的候选解决方案提供了独特的机会，因为科学家、政策制定者和学生从未像今天这样接触到更多的电子信息。这项研究的目标是开发新的文本挖掘方法，这些方法与专家目前用来解决矛盾和冗余证据的手动过程一致。即使对人类来说，发现和合成都是困难的活动，因此团队计划制定一项社会技术战略来实现这一目标。这项研究包括对北卡罗来纳大学和三角研究园区的教师、政策制定者和学生组成的不同网络的人工发现和综合行为的纵向研究。主要的工作将是改进自然语言处理方法，以自动识别概念和关系，检测蕴含和释义，并生成多文档摘要。最后，将进行一系列定性和定量研究，准确反映文本挖掘方法在发现和合成活动中的帮助程度。该项目将推进语言处理方法，以检测概念和关系，识别释义和蕴涵，并生成多个文件摘要；为自然语言社区提供反映多样化和现实信息需求的黄金标准集合；培训下一代科学家，探索跨学科的复杂研究；促进发现的人类方面？通过一个赞助的研讨会。本项目提出的文本挖掘的社会技术解决方案将确保对后续的文本挖掘理论和工具产生广泛的影响。该项目将通过使专家能够跟踪学科之间的联系来加速科学发现，并通过减少解决学科内看似多余和矛盾的证据所需的时间来加快政策制定。让环境保护局和塞西尔·G·谢普斯卫生服务中心的政策倡导者参与进来，将确保这项研究产生的理论和技术与发现和综合发生的复杂环境保持一致。索赔跳跃器将加快他们现有的政策努力，但更重要的是，来自该项目的工具将使手动方法无法实现的研究成为可能。

Vast quantities of electronic information provide a unique opportunity for scientists identify candidate solutions for grand challenges as scientists, policy makers, and students have never had access to more electronic information than they do today. The goal in this research is to develop new text mining methods that are consistent with the manual processes that experts currently used to resolve contradictory and redundant evidence. Both discovery and synthesis are difficult activities even for people, so the team plans a socio-technical strategy to achieve this goal. This study includes a longitudinal study of manual discovery and synthesis behaviors of a diverse network of faculty, policy makers, and students from UNC and the Research Triangle Park. The majority of effort will be to advance natural language processing methods that automatically identify concepts and relationships, detect entailment and paraphrasing, and generate multi-document summaries. Lastly, a series of qualitative and quantitative studies that accurately reflect the degree to which text mining methods assist in discovery and synthesis activities will be conducted. This project will advance language processing methods that detect concepts and relationships, recognize paraphrases and entailment, and generate multiple documents summaries; provide the natural language community with a collection of gold standards that reflect diverse and realistic information needs; train the next generation of scientists to explore complex research that span disciplines; promote the ?human side of discovery? via a sponsored workshop. The socio-technical solution to text mining proposed in this project will ensure broad impact of the subsequent text mining theory and tools. This project will accelerate scientific discovery by enabling experts to follow connections between disciplines; and accelerate policy development by reducing the time required to resolve seemingly redundant and contradictory evidence within a discipline. Involving policy champions from the Environmental Protection Agency and the Cecil G. Sheps Center for Health Services will ensure that the theory and technology produced from this research are consistent with the complex environment in which discovery and synthesis takes place. Claim Jumper will accelerate their existing policy efforts, but more importantly, tools from this project will enable studies that are not feasible with manual methods.