source: orange/docs/widgets/rst/classify/naivebayes.rst @ 11359:8d54e79aa135

Revision 11359:8d54e79aa135, 3.0 KB checked in by Ales Erjavec <ales.erjavec@…>, 14 months ago (diff)

Cleanup of 'Widget catalog' documentation.

Fixed rst text formating, replaced dead hardcoded reference links (now using
:ref:), etc.

RevLine 
[11050]1.. _Naive Bayes:
2
3Naive Bayesian Learner
4======================
5
6.. image:: ../icons/NaiveBayes.png
7
8Naive Bayesian Learner
9
10Signals
11-------
12
13Inputs:
14
15
16   - Examples (ExampleTable)
17      A table with training examples
18
19
20Outputs:
21
22   - Learner
[11359]23      The naive Bayesian learning algorithm with settings as specified in
24      the dialog.
[11050]25
26   - Naive Bayesian Classifier
27      Trained classifier (a subtype of Classifier)
28
29
[11359]30Signal :code:`Naive Bayesian Classifier` sends data only if the learning
31data (signal :code:`Examples` is present.
[11050]32
33Description
34-----------
35
36This widget provides a graphical interface to the Naive Bayesian classifier.
37
[11359]38As all widgets for classification, this widget provides a learner and
39classifier on the output. Learner is a learning algorithm with settings
40as specified by the user. It can be fed into widgets for testing learners,
41for instance :ref:`Test Learners`. Classifier is a Naive Bayesian Classifier
42(a subtype of a general classifier), built from the training examples on the
43input. If examples are not given, there is no classifier on the output.
[11050]44
45.. image:: images/NaiveBayes.png
46   :alt: NaiveBayes Widget
47
[11359]48Learner can be given a name under which it will appear in, say,
49:ref:`Test Learners`. The default name is "Naive Bayes".
[11050]50
[11359]51Next come the probability estimators. :obj:`Prior` sets the method used for
52estimating prior class probabilities from the data. You can use either
53:obj:`Relative frequency` or the :obj:`Laplace estimate`.
54:obj:`Conditional (for discrete)` sets the method for estimating conditional
55probabilities, besides the above two, conditional probabilities can be
56estimated using the :obj:`m-estimate`; in this case the value of m should be
57given as the :obj:`Parameter for m-estimate`. By setting it to
58:obj:`<same as above>` the classifier will use the same method as for
59estimating prior probabilities.
[11050]60
[11359]61Conditional probabilities for continuous attributes are estimated using
62LOESS. :obj:`Size of LOESS window` sets the proportion of points in the
63window; higher numbers mean more smoothing.
64:obj:`LOESS sample points` sets the number of points in which the function
65is sampled.
[11050]66
[11359]67If the class is binary, the classification accuracy may be increased
68considerably by letting the learner find the optimal classification
69threshold (option :obj:`Adjust threshold`). The threshold is computed from
70the training data. If left unchecked, the usual threshold of 0.5 is used.
[11050]71
[11359]72When you change one or more settings, you need to push :obj:`Apply`;
73this will put the new learner on the output and, if the training examples
74are given, construct a new classifier and output it as well.
[11050]75
76
77Examples
78--------
79
[11359]80There are two typical uses of this widget. First, you may want to induce
81the model and check what it looks like in a :ref:`Nomogram`.
[11050]82
83.. image:: images/NaiveBayes-SchemaClassifier.png
84   :alt: Naive Bayesian Classifier - Schema with a Classifier
85
[11359]86The second schema compares the results of Naive Bayesian learner with
87another learner, a C4.5 tree.
[11050]88
89.. image:: images/C4.5-SchemaLearner.png
90   :alt: Naive Bayesian Classifier - Schema with a Learner
Note: See TracBrowser for help on using the repository browser.