Context for Data Visualization

The process of data visualization often imposes a loss of context which would help the interpretation of that data and its visual representation. In communicative data visualization this context is often added back by editors such as journalists who provide messages or “chatter” to data graphics. But the degree of manual effort required to create thoughtfully messaged visualizations is considerable and does not scale well to the deluge of public data available. In this work we describe the design of a prototype system called Contextifier that automatically produces custom annotated visualizations of stock data. Contextifier’s algorithm for choosing messages is informed by the study of journalistic narrative visualizations and takes into account visual salience, contextual relevance, and a detection of key events in the company’s history.

  • J. Hullman, N. Diakopoulos, E. Adar. Contextifier: Automatic Generation of Annotated Stock VisualizationsConference on Human Factors in Computing Systems (CHI). May, 2013. [PDF]
  • T. Gao, J. Hullman, E. Adar, B. Hecht, N. Diakopoulos. NewsViews: An Automated Pipeline for Creating Custom Geovisualizations for News. Proc. Conference on Human Factors in Computing Systems (CHI). May, 2014. [PDF]