Günnemann, NikouSaake, GunterHenrich, AndreasLehner, WolfgangNeumann, ThomasKöppen, Veit2018-10-242018-10-242013978-3-88579-610-7https://dl.gi.de/handle/20.500.12116/17438Recent developments in web technologies like Web 2.0 have led to the generation of massive amounts of data. The rapid growth of data makes knowledge extraction and trend prediction a challenging task. A recent approach for the unsupervised analysis of text corpora is dynamic topic mining. While there is a growing interest in using this technique, interactive analysis systems for dynamic topic mining are still in an early stage. In this paper we present D-VITA, an interactive text analysis system that exploits dynamic topic mining to detect the latent topic structure and topic dynamics in a collection of documents. D-VITA supports end-users in understanding and exploiting the topic mining results, in visualizing the topic dynamics within document collections, and in browsing of documents based on shared topics. We present an application case for a scientific community that uses an instance of D-VITA for trend analysis in their data sources.enText/Conference Paper1617-5468