Linking unstructured evidence to structured observations

Manuela Waldner*, Dieter Schmalstieg, Thomas Geymayer, Michael Sedlmair

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

Many professionals, like journalists, writers, or consultants, need to acquire information from various sources, make sense of this unstructured evidence, structure their observations, and finally create and deliver their product, such as a report or a presentation. In formative interviews, we found that tools allowing structuring of observations are often disconnected from the corresponding evidence. Therefore, we designed a sensemaking environment with a flexible observation graph that visually ties together evidence in unstructured documents with the user’s structured knowledge. This is achieved through bi-directional deep links between highlighted document portions and nodes in the observation graph. In a controlled study, we compared users’ sensemaking strategies using either the observation graph or a simple text editor on a large display. Results show that the observation graph represents a holistic, compact representation of users’ observations, which can be linked to unstructured evidence on demand. In contrast, users taking textual notes required much more display space to spatially organize source documents containing unstructured evidence. This implies that spatial organization is a powerful strategy to structure observations even if the available space is limited
Original languageEnglish
Pages (from-to)47-65
Number of pages19
JournalInformation Visualization
Volume20
Issue number1
Early online date14 Jan 2021
DOIs
Publication statusPublished - Jan 2021

Keywords

  • concept map
  • Mind map
  • observation graph
  • sensemaking
  • visual links

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Fingerprint Dive into the research topics of 'Linking unstructured evidence to structured observations'. Together they form a unique fingerprint.

Cite this