The Geospatial Web

Generated by orthographically projecting and tiling semantic maps (= visual representations of semantic information spaces based on a landscape metaphor), knowledge planets allow visualizing massive amounts of textual data. At the time of map generation, the planet’s topology is determined by the content of the knowledge base. The peaks of the virtual landscape indicate abundant coverage on a particular topic, whereas valleys represent sparsely populated parts of the information space.

Prototype (requires OpenGL support)

www.ecoresearch.net/climate (Media Watch on Climate Change)
www.ecoresearch.net/climate/knowledge-planet (Knowledge Planet)

Controls

• Pan: Left mouse button, click and drag (all directions)
• Zoom: Use the scroll wheel on the mouse
• Tilt: Right mouse button click & drag (up and down), or use "Page Up" and "Page Down" on the keyboard.
• Rotate: Right mouse button click & drag (left and right; crossing the top and bottom half of the screen while rotating will change direction).
• Reset Heading: "N"; Reset all: "R"

Screenshot

Background

Building upon the work of Sabol et al. (2002), the current software development is being undertaken by Syed Kamran Ali Ahmad, Alexander Hubmann-Haidvogel and Dimitri Zibold as part of the IDIOM and RAVEN research projects. Initially, the document set is pre-clustered using hierarchical agglomerative clustering (Jain et al. 1999), randomly distributing the cluster centroids in the viewing rectangle. The documents belonging to each cluster, as determined by the pre-clustering, are then placed in circles around each centroid. The arrangement is fine-tuned using linear iteration force-directed placement adapted from Chalmers (1996). The result resembles a contour map of islands.

Fortunately, algorithms based on force models easily generalize to the knowledge planets’ spherical geometries. The IDIOM (Information Diffusion across Interactive Online Media; www.idiom.at) research project extends and refines the original thematic mapping component to improve throughput and scalability, generate layered thematic maps, and provide a Web Map Service (WMS) that serves these maps as image tiles for various geobrowsing platforms (Scharl 2007).

The NASA World Wind screenshot above shows the prototype as of May 2008. The transition from two-dimensional thematic maps to three-dimensional knowledge planets poses a number of conceptual and technical challenges – the initial arrangement of major concepts, for example, which should be guided by domain ontologies. Users will also expect a consistent experience when rotating the planet. This requires a seamless flow of concepts when crossing the planet’s 0° meridian line. The same principle applies to zooming operations. Analogous to Landsat-7 data, multiple layers of thematic maps in different resolutions and with appropriate sets of captions have to be synchronized with each other.

References

Andrews, K., Guetl, C., Moser, J., Sabol, V. and Lackner, W. (2001). “Search Result Visualization with xFIND”, Second International Workshop on User Interfaces to Data Intensive Systems (UIDIS 2001). Zurich: IEEE Press, 50–58.

Chalmers, M. (1996). “A Linear Iteration Time Layout Algorithm for Visualizing High-Dimensional Data”, 7th Conference on Visualization. San Francisco, CA: IEEE Computer Society, 127–132.

Jain, A.K., Murty, M.N. and Flynn, P.J. (1999). “Data Clustering: A Review”, ACM Computing Surveys, 31(3): 264–323.

Sabol, V., Kienreich, W., Granitzer, M., Becker, J., Tochtermann, K. and Andrews, K. (2002). “Applications of a Lightweight, Web-based Retrieval, Clustering, and Visualisation Framework”, 4th International Conference on Practical Aspects of Knowledge Management (LNCS, Vol. 2569). Eds. D. Karagiannis and U. Reimer. Berlin: Springer, 359–368.

Scharl, A. (2007). “Towards the Geospatial Web: Media Platforms for Managing Geotagged Knowledge Repositories”, The Geospatial Web – How Geo-Browsers, Social Software and the Web 2.0 are Shaping the Network Society. Eds. A. Scharl and K. Tochtermann. London: Springer. 3-14.