SGER: Intelligent Patent Analysis and Visualization

技术创新的衡量和评估吸引了许多行为者的兴趣。正如国家科学和技术统计出版物所反映的那样，近几十年来，在宏观和微观经济分析以及政策使用方面，对可靠的创新指标的需求稳步增加。拟议研究的目标是利用数据/文本分析和可视化技术的进步来支持知识发现，并根据(NSF)项目资金和(USPTO)专利数据库确定知识转移。由于发明产出的指标大多以专利为基础，因此需要有效和创新的方法和框架来分析存储在专利数据库中的与技术创新有关的信息。关于国家科学基金资助趋势的信息可以从对国家科学基金奖摘要和其他相关数据的分析中得出。研究面临的挑战是：使用奖项和专利数据来近似科学和技术发展的有效性存在不确定性，难以直观地呈现分析结果，以及准确识别NSF奖项和专利数据库中的新兴主题和概念。这一探索性项目旨在研究技术问题和涉及的基本假设。采用的主要信息和计算机科学技术包括：基于文本的语言分析、2D/3D知识地图可视化、引文网络分析以及时间模式的图形显示和分析。目前的测试平台与美国纳米科学和技术研究的分析以及http://ai.bpa.arizona.edu/.上描述的其他研究有关

The measurement and assessment of technological innovation has attracted the interest of many actors. As reflected in national science and technology statistical publications, the demand for reliable innovation indicators has steadily increased over recent decades, both for macro and micro-economic analysis, and for policy use. The goal of the proposed research is to leverage advances in data/text analysis and visualization techniques to support knowledge discovery and to identify knowledge transfer based on (NSF) project funding and (USPTO) patent databases. As indicators of inventive output are mostly based on patents, efficient and innovative methodologies and frameworks for the analysis of the information related to technological innovation stored in patent databases are needed. Information about trends in NSF funding can be derived from the analysis of the NSF award abstracts and other relevant data. The research challenges are: uncertainty about the validity of using the award and patent data to approximate science and technology development, difficulty in intuitive presentation of analysis result, and precise identification of emerging topics and concepts in the NSF award and patent databases. This exploratory project aims to examine both technical issues and the fundamental hypotheses involved. Major information and computer science techniques adopted include: text-based linguistic analysis, 2D/3D knowledge map visualization, citation network analysis, and graphical display and analysis of temporal patterns. The current testbed is related to analysis of nano science and technology research in USA and other research described at http://ai.bpa.arizona.edu/.