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

Revision 11778:ecd4beec2099, 2.2 KB checked in by Ales Erjavec <ales.erjavec@…>, 5 months ago (diff)

Use new SVG icons in the widget documentation.

Line 
1.. _Sieve Diagram:
2
3Sieve Diagram
4=============
5
6.. image:: ../../../../Orange/OrangeWidgets/Visualize/icons/SieveDiagram.svg
7
8Plots a sieve diagram for a pair of attributes.
9
10Signals
11-------
12
13Inputs:
14   - Examples (ExampleTable)
15      Input data set.
16
17
18Outputs:
19   - None
20
21
22Description
23-----------
24
25A sieve diagram is a graphical method for visualizing the frequencies in a
26two-way contingency table and comparing them to the expected frequencies under
27assumtion of independence. The sieve diagram was proposed by Riedwyl and
28Schüpbach in a technical report in 1983 and later called a parquet diagram
29([Riedwy1994]_). In this display the area of each rectangle is proportional to
30expected frequency and observed frequency is shown by the number of squares in
31each rectangle. The difference between observed and expected frequency
32(proportional to standard Pearson residual) appears as the density of shading,
33using color to indicate whether the deviation from independence is positive
34(blue) or negative (red).
35
36The snapshot below shows a sieve diagram for Titanic data set and attributes
37sex and survived (the later is actually a class attribute in this data set).
38The plot shows that the two variables are highly associated, as there are
39substantial differences between observed and expected frequencies in all of the
40four quadrants. For example and as highlighted in a balloon, the chance for
41not surviving the accident was for female passengers much lower than expected
42(0.05 vs. 0.14).
43
44.. image:: images/SieveDiagram-Titanic.png
45
46Orange can help to identify pairs of attributes with interesting associations.
47Such attribute pairs are upon request (:obj:`Calculate Chi Squares`) listed
48in :obj:`Interesting attribute pair`. As it turns out, the most interesting
49attribute pair in Titanic data set is indeed the one we show in the above
50snapshot. For a contrast, the sieve diagram of the least interesting pair
51(age vs. survival) is shown below.
52
53.. image:: images/SieveDiagram-Titanic-age-survived.png
54
55References
56----------
57
58.. [Riedwy1994] Riedwyl, H., and Schüpbach, M. (1994). Parquet diagram to plot
59   contingency tables. In  Softstat '93: Advances in Statistical Software,
60   F. Faulbaum (Ed.). New York: Gustav Fischer, 293-299.
Note: See TracBrowser for help on using the repository browser.