Incremental and Scalable Computation of Dynamic Topography Information Landscapes

Dynamic topography information landscapes are capable of visualizing longitudinal changes in large document repositories. Resembling tectonic processes in the natural world, dynamic rendering reflects both long-term trends and short-term fluctuations in such repositories. To visualize the rise and decay of topics, the mapping algorithm elevates and lowers related sets of concentric contour lines. Acknowledging the growing number of documents to be processed by stateof-the-art Web intelligence applications, we present a scalable, incremental approach for generating such landscapes. The processing pipeline includes a number of sequential tasks, from crawling, filtering and pre-processing Web content to projecting, labeling and rendering the aggregated information. Processing steps central to incremental processing are found in the projection stage which consists of document clustering, cluster force-directed placement, and fast document positioning. We introduce two different positioning methods and compare them in an incremental setting using two different quality measures. The evaluation is performed on a set of approximately 5000 documents taken from the environmental blog sample of the Media Watch on Climate Change (www.ecoresearch.net/climate), a Web content aggregator about climate change and related environmental issues that serves static versions of the information landscapes presented in this paper as part of a multiple coordinated view representation.