Advanced annotation and text models for visual analytics

我们生活在一个信息过载的时代，在商用飞机安全和维护领域，每天以自然语言的形式产生大量的文本，这些文本可能为安全分析师提供至关重要的信息。然而，自由形式的自然语言文本中的信息很难被机器处理，我们需要有效的人机交互工具来分析文本数据。许多不同学科的分析师需要计算工具来帮助他们浏览大量文本数据集，并找到特定的信息。我们的项目将产生一个改进的可视化分析软件系统，专门为大型文本数据集的可视化设计。我们将把重点放在这项研究的一个特定的公共安全应用上，这将使涉及商用飞机安全研究的行业受益，特别是我们在这个项目中的工业合作伙伴：AeroInfo Systems和波音公司。我们的研究将在可视化分析领域产生加拿大的存在，我们的软件可以扩展到支持从财务报告分析和刑事案件到对加拿大至关重要的国家安全和国防的各个层面的调查分析。

We live in an age of information overload, and in the area of commercial airplane safety and maintenance a vast amount of text is produced every day in the form of natural language, which may yield information vital for safety analysts. However, information in free-form natural language text is difficult to process by machine, and we need effective tools for human-computer interaction for the analysis of text data. Analysts in many different disciplines need computational tools that help them navigate through massive text data sets and find specific nuggets of information. Our project will produce an improved Visual Analytics software system specifically designed for visualization in large text data sets. We will focus on a particular public safety application of this research that will benefit industries involved in commercial airplane safety studies, in particular our industrial partner in this project: AeroInfo Systems and The Boeing Company. Our research will produce a Canadian presence in the field of visual analytics and our software can be extended to support investigative analysts at all levels from financial report analysis and criminal cases to national security and defence which are of vital importance to Canada.