source: orange/docs/widgets/rst/visualize/parallelcoordinates.rst @ 11778:ecd4beec2099

Revision 11778:ecd4beec2099, 3.6 KB checked in by Ales Erjavec <ales.erjavec@…>, 5 months ago (diff)

Use new SVG icons in the widget documentation.

1.. _Parallel Coordinates:
3Parallel Coordinates
6.. image:: ../../../../Orange/OrangeWidgets/Visualize/icons/ParallelCoordinates.svg
8Parallel Coordinates visualization with some explorative data analysis and
9intelligent data visualization enhancements.
15   - Examples (ExampleTable)
16      Input data set.
17   - Example Subset (ExampleTable)
18      A subset of data instances from Examples.
19   - Attribute Selection List
20      List of attributes to be shown in the visualization.
24   - Selected Examples (ExampleTable)
25      A subset of examples that user has manually selected from the
26      scatterplot.
27   - Unselected Examples (ExampleTable)
28      All other examples (examples not included in the user's selection).
29   - Attribute Selection List
30      List of attributes used in the visualization.
36Parallel Coordinates is a multidimensional data visualization technique. Each
37attribute is represented in a vertical line, where the maximum and minimum
38values of that dimension are scaled to the upper and lower points on these
39vertical lines. For N visualized attributes, N-1 lines connected to each
40vertical line at the appropriate dimensional value represent an N-dimensional
41point. The snapshot shown below displays data from the Iris data set, with
42the data instance closest to the cursor being highlighted. In Iris data set,
43the instances are labeled with one of the three distinct classes, depicted with
44colored lines in the visualization (red, green, blue).
46.. image:: images/ParallelCoordinates-Iris.png
48The :obj:`Main` tab allows the user to choose the subset of attributes to be
49displayed in the visualization. In case of a class-labeled data set, only one
50class (vs. the others) may be exposed by selecting it from :obj:`Target class`.
51Especially with data sets that include many attributes,
52:obj:`Optimization Dialog` may help to find interesting projections. Currently,
53this is decided based on a correlation between neighboring attributes in the
54visualization, where the target s to find visualizations with the highest sum
55of the absolute value of correlations between neighboring attributes. Snapshot
56below shows such a visualization which uses five attributes and plots the data
57set from functional genomics (****).
59.. image:: images/ParallelCoordinates-Optimization.png
61The :obj:`Settings` tab is used to control several aspects on visualization.
62:obj:`Jittering` may be useful when the data includes instances which share
63many of the attribute values. The graph can be annotated by displaying minimal
64(bottom) and maximal (top) values of the attributes
65(:obj:`Show attribute values`). For polygon of each of the data instances can
66be converted to a spline (:obj:`Show splines`). :obj:`Global value scaling`
67would make the scale for each of the attributes equal by finding the extreme
68values across the attributes in the display. That could be useful in a number
69of applications, such as, for instance, those from functional genomics (the
70snapshot shown below). :obj:`Line tracking` can highlight the polygon of an
71instance closest to the mouse pointer (see the topmost snapshot on this page).
72The option :obj:`Hide pure examples` would draw the data instance polygons from
73left to right, stopping at the attribute where at a distinct point all the
74instances would belong to a single class. The setting in :obj:`Statistics` box
75toggles the drawing of an average or median trajectory. Information on
76correlation between the neighboring attributes may also be displayed
77(:obj:`Between-axis labels`).
79.. image:: images/ParallelCoordinates-Settings.png
Note: See TracBrowser for help on using the repository browser.