BIOMEDICAL (APPLIED/EXPLORATORY)

该项目将使NIH的分析师和政策制定者能够通过基于Web的原型可视化工具，使用主题建模和大规模网络可视化来探索和理解数千至数百万份文件中的关键信息，例如赠款申请和生物医学研究数据。第一阶段将侧重于：(1)开发一个有效的网络数据模型；(2)创建一个生成有意义的可视化的工具；(3)利用专题专家的帮助，为非技术用户开发一个有效的用户界面；(4)为第二阶段确定改进的机会和领域。该工具将通过使用混合的客户端-服务器可视化架构扩展到数百万个文档，该架构(1)在服务器端生成静态可视图像，(2)在客户端使用基于浏览器的技术提供元数据、搜索和过滤能力。该工具只是ChalkLabs正在开发的高级视觉分析平台的一个应用程序。

This project will enable analysts and policy-makers at NIH to explore and understand key information in thousands to millions of documents such as grant applications and biomedical research data through a prototypical Web-based visualization tool using Topics modeling and large-scale network visualization. Phase I will be focused on: (1) Developing an efficient network data model; (2) Creating a tool that generates meaningful visualizations; (3) Developing an effective user interface for non-technical users using the help of subject-matter experts; and (4) Identifying opportunities and areas of improvement for Phase II. The tool will scale to millions of documents by using hybrid client-server visualization architecture that (1) generates static visual images on the server-side and (2) provides metadata, search, and filtering capabilities using browser-based technologies on the client-side. The tool is only one application of an advanced visual-analytics platform being developed by ChalkLabs.