Context stamp: a topic-based content abstraction for visual concordance analysis.

Concordance analysis supports users in studying how terms are used in a document vs. another by investigating their usage contexts. As current approaches usually present a large set of contexts in their full text form or as a large frequency-based word cloud, they still require a lot of effort from users to make sense of the underlying complex and dynamic semantic dimensions of contexts. To address this limitation, we propose Context Stamp as a visual representation of the gist of a term's usage contexts. To abstract away the textual details and yet retain the core facets of a term's contexts for visualization, we blend a statistical topic modeling method with a combination of the treemaps and Seesoft visual metaphors. This paper provides a high level description of the text analysis method and outlines the visual design of Context Stamps.