source: orange/docs/widgets/rst/evaluate/liftcurve.rst @ 11778:ecd4beec2099

Revision 11778:ecd4beec2099, 2.3 KB checked in by Ales Erjavec <ales.erjavec@…>, 5 months ago (diff)

Use new SVG icons in the widget documentation.

Line 
1.. _Lift Curve:
2
3Lift Curve
4==========
5
6.. image:: ../../../../Orange/OrangeWidgets/Evaluate/icons/LiftCurve.svg
7
8Shows the lift curves and analyzes them.
9
10Signals
11-------
12
13Inputs:
14   - Evaluation Results (orngTest.ExperimentResults)
15      Results of classifiers' tests on data
16
17
18Outputs:
19   - None
20
21Description
22-----------
23
24Lift curves show the relation between the number of instances which were
25predicted positive and those of them that are indeed positive. This type of
26curve is often used in segmenting the population, e.g., plotting the number
27of responding customers against the number of all customers contacted. Given
28the costs of false positives and false negatives, it can also determine the
29optimal classifier and threshold.
30
31.. image:: images/LiftCurve.png
32
33Option :obj:`Target class` chooses the positive class. In case there are
34more than two classes, the widget considers all other classes as a single,
35negative class.
36
37If the test results contain more than one classifier, the user can choose
38which curves she or he wants to see plotted. :obj:`Show convex lift hull`
39plots a convex hull over lift curves for all classifiers. The curve thus
40shows the optimal classifier (or combination thereof) for each desired TP/P
41rate. The diagonal line represents the behaviour of a random classifier.
42
43The user can specify the cost of false positives and false negatives, and
44the prior target class probability. :obj:`Compute from Data` sets it to the
45proportion of examples of this class in the data. The black line in the
46graph, which corresponds to the right-hand axis, gives the total cost for
47each P ration for the optimal classifier among those selected in the list
48box on the left. The minimum is labelled by the optimal classifier at that
49point and the related cost.
50
51The widget allows setting costs from 1 to 1000. The units are not important,
52as are not the magnitudes. What matters is the relation between the two
53costs, so setting them to 100 and 200 will give the same result as 400 and 800.
54
55Example
56-------
57
58At the moment, the only widget which give the right type of the signal
59needed by the Lift Curve is :ref:`Test Learners`. The Lift Curve will hence
60always follow Test Learners and, since it has no outputs, no other widgets
61follow it. Here is a typical example.
62
63.. image:: images/ROCLiftCalibration-Schema.png
Note: See TracBrowser for help on using the repository browser.