 |
 |
|
Evènement pour le groupe BDVISU
| Date | 2007-06-01 09:30-10:30 |
| Titre | Grouse: Feature-Based, Steerable Graph Hierarchy |
| Résumé | Grouse is a feature-based approach to steerable
exploration of a graph and an associated hierarchy.
Steerability allows exploration to begin immediately,
rather than requiring a costly layout of the entire
graph as an initial
step. In a feature-based approach, the subgraph inside
a metanode of the graph hierarchy is laid out with a
well-chosen algorithm appropriate for its topological
structure. Grouse preserves the input hierarchy, which
provides meaningful information to the user when its
metanodes correspond to features of interest. When a
metanode in the hierarchy is opened, a limited number
of metanodes are laid out again along the path between
the opened node and the root. We demonstrate the
effectiveness of Grouse on datasets from IMDB, the
Internet Movie Database, where nodes are actors and
cliques represent movies. The combination of
feature-based layout and limited relayout
computation does not fragment features in the
hierarchy and improves the number of levels in the
hierarchy that
can be seen at once over previous approaches.
|
| Lieu | 075 |
| Orateur | Daniel Archambault |
| Email | archam@cs.ubc.ca |
| Url | University of British Columbia, Canada |
1 document lié: grouseEvisTalk.pdf
RetourRetour à l'index
| |