Datasets:
contextualizing-scientific-claims
/
extracted_captions
/goyalEffectsSensemakingTranslucence2016.json
{"CAPTION FIG1.png": "'Figure 1: The Document Space showing (clockwise, from top-left) the directory of crime case documents, a tabbed reader panel for reading case documents, a visual graph of connections based on common entities in the dataset, a map to identify locations of crimes and events, and a timeline to track events.\\n\\n'", "CAPTION FIG2.png": "'Figure 2: The Analysis Space showing (clockwise, from top-left) the chat for explicit sharing, connected sticklies for Implicit Sharing, hypothesis window in the middle with columns to add new Hypotheses, confirming evidence, and disconfirming evidence for explicit Hypothesis Tracking, and Suspect Visualization at the bottom with 4 Avatars: Dennis Rathbone. Marilyn Stokes, Steve Gramming, and Lousie for Suspect Tracking. Note: Chat, Visualization, Sticky, and Hypothesis Window have been magnified to improve readability.\\n\\n'", "CAPTION FIG3.png": "'Figure 3: Sample sensemaking trajectory 1. Sensemaking-Translucence reminds users to consider suspects by showing empty Avatars at the start 2. Avatars are automatically populated by names detected from implicitly shared stickies. 3, 4 & 5. Avatars show distribution of name-reference by getting darker for names mentioned in stickies, chat and hypothesis window. The last mentioned suspect in hypothesis window is marked red 6. With use, visualization depicts distribution of attention at suspect level, based on explicit mentions. Note: Chat, Visualization, Sticky, and Hypothesis Window have been magnified to improve readability in 2, 3, 4 and 5. 1 and 6 represent non-magnified versions of the Sensemaking Translucence interface.\\n\\n'", "CAPTION FIG4.png": "'Figure 4. Serial killer identification and number of correct clues identified by interface condition.\\n\\n'", "CAPTION FIG5.png": "'Figure S. Perception of interface usefulness by interface condition\\n\\n'", "CAPTION FIG6.png": "'Figure 6: Self-reported workload and team experience by interface condition.\\n\\n'"} |