Data Cubes and Zoom Graphs
Data Cubes and Zoom Graphs Chris Gray [email protected]
Data Cubes and Zoom Graphs – p. 1/14
Intro Zoom graphs Data cubes Examples
Data Cubes and Zoom Graphs – p. 2/14
Polaris Review Database exploration. A formal specification language for describing visualizations of database tables. An IDE for generating the specifications. A database query generation system to get the data to be visualized.
Data Cubes and Zoom Graphs – p. 3/14
Big Picture
Data Cubes and Zoom Graphs – p. 4/14
Zoom Graphs Like a space/scale diagram, but discrete. Going from node to node via an edge represents a zoom.
Data Cubes and Zoom Graphs – p. 5/14
Differences from ZUIs Not intended as a desktop replacement. Multiple dimensions.
Data Cubes and Zoom Graphs – p. 6/14
n-dimensional cubes Dimensions and measures. One dimension for each variable. Each variable has a “measure” – its magnitude. Each dimension can be at a different level of detail. Ordinal data must be given an ordering.
Data Cubes and Zoom Graphs – p. 7/14
Lattice ≤ relation between some cubes. Defined by a higher level of detail in one dimension.
Data Cubes and Zoom Graphs – p. 8/14
Visual Encodings Possible to specify visual encodings completely in the graphs. Color Quantitative ramp, qualitative. Size Height, width, or both. Shape Text, point, line, or polygon. Data Cubes and Zoom Graphs – p. 9/14
Layers As user zooms in, more layers emerge. Labels appear when user is close enough. Aggregated data become separated.
Data Cubes and Zoom Graphs – p. 10/14
Example
Data Cubes and Zoom Graphs – p. 11/14
Problems Continuous zooms are hard to do with this abstraction. Design decision. Arbitrary zooms are not possible.
Data Cubes and Zoom Graphs – p. 12/14
Conclusions Data cubes Zoom graphs
Data Cubes and Zoom Graphs – p. 13/14
Questions?
Data Cubes and Zoom Graphs – p. 14/14
References [1] Chris Stolte, Diane Tang and Pat Hanrahan, “Multiscale Visualization Using Data Cubes”, Proceedings of the Eighth IEEE Symposium on Information Visualization, October 2002. [2] Chris Stolte, Diane Tang and Pat Hanrahan, “Polaris”, http://graphics.stanford.edu/projects/polaris/
14-1
Data Cubes and Zoom Graphs – p. 1/14
Intro Zoom graphs Data cubes Examples
Data Cubes and Zoom Graphs – p. 2/14
Polaris Review Database exploration. A formal specification language for describing visualizations of database tables. An IDE for generating the specifications. A database query generation system to get the data to be visualized.
Data Cubes and Zoom Graphs – p. 3/14
Big Picture
Data Cubes and Zoom Graphs – p. 4/14
Zoom Graphs Like a space/scale diagram, but discrete. Going from node to node via an edge represents a zoom.
Data Cubes and Zoom Graphs – p. 5/14
Differences from ZUIs Not intended as a desktop replacement. Multiple dimensions.
Data Cubes and Zoom Graphs – p. 6/14
n-dimensional cubes Dimensions and measures. One dimension for each variable. Each variable has a “measure” – its magnitude. Each dimension can be at a different level of detail. Ordinal data must be given an ordering.
Data Cubes and Zoom Graphs – p. 7/14
Lattice ≤ relation between some cubes. Defined by a higher level of detail in one dimension.
Data Cubes and Zoom Graphs – p. 8/14
Visual Encodings Possible to specify visual encodings completely in the graphs. Color Quantitative ramp, qualitative. Size Height, width, or both. Shape Text, point, line, or polygon. Data Cubes and Zoom Graphs – p. 9/14
Layers As user zooms in, more layers emerge. Labels appear when user is close enough. Aggregated data become separated.
Data Cubes and Zoom Graphs – p. 10/14
Example
Data Cubes and Zoom Graphs – p. 11/14
Problems Continuous zooms are hard to do with this abstraction. Design decision. Arbitrary zooms are not possible.
Data Cubes and Zoom Graphs – p. 12/14
Conclusions Data cubes Zoom graphs
Data Cubes and Zoom Graphs – p. 13/14
Questions?
Data Cubes and Zoom Graphs – p. 14/14
References [1] Chris Stolte, Diane Tang and Pat Hanrahan, “Multiscale Visualization Using Data Cubes”, Proceedings of the Eighth IEEE Symposium on Information Visualization, October 2002. [2] Chris Stolte, Diane Tang and Pat Hanrahan, “Polaris”, http://graphics.stanford.edu/projects/polaris/
14-1
