posted on 2012-05-01, 00:00authored byJacob Eisenstein, Duen Horng Chau, Aniket KitturAniket Kittur, Eric P Xing
<p>Existing methods for searching and exploring large document collections focus on surface-level matches to user queries, ignoring higher-level semantic structure. In this paper we show how topic modeling — a technique for identifying latent themes across a large collection of documents — can support semantic exploration. We present TopicViz: an interactive environment which combines traditional search and citation-graph exploration with a force-directed layout that links documents to the latent themes discovered by the topic model. We describe usage scenarios in which TopicViz supports rapid sensemaking on large document collections.</p>