1AI::Categorizer::LearneUrs(e3r)Contributed Perl DocumentAaIt:i:oCnategorizer::Learner(3)
2
3
4

NAME

6       AI::Categorizer::Learner - Abstract Machine Learner Class
7

SYNOPSIS

9        use AI::Categorizer::Learner::NaiveBayes;  # Or other subclass
10
11        # Here $k is an AI::Categorizer::KnowledgeSet object
12
13        my $nb = new AI::Categorizer::Learner::NaiveBayes(...parameters...);
14        $nb->train(knowledge_set => $k);
15        $nb->save_state('filename');
16
17        ... time passes ...
18
19        $nb = AI::Categorizer::Learner::NaiveBayes->restore_state('filename');
20        my $c = new AI::Categorizer::Collection::Files( path => ... );
21        while (my $document = $c->next) {
22          my $hypothesis = $nb->categorize($document);
23          print "Best assigned category: ", $hypothesis->best_category, "\n";
24          print "All assigned categories: ", join(', ', $hypothesis->categories), "\n";
25        }
26

DESCRIPTION

28       The "AI::Categorizer::Learner" class is an abstract class that will
29       never actually be directly used in your code.  Instead, you will use a
30       subclass like "AI::Categorizer::Learner::NaiveBayes" which implements
31       an actual machine learning algorithm.
32
33       The general description of the Learner interface is documented here.
34

METHODS

36       new()
37           Creates a new Learner and returns it.  Accepts the following
38           parameters:
39
40           knowledge_set
41               A Knowledge Set that will be used by default during the train()
42               method.
43
44           verbose
45               If true, the Learner will display some diagnostic output while
46               training and categorizing documents.
47
48       train()
49       train(knowledge_set => $k)
50           Trains the categorizer.  This prepares it for later use in
51           categorizing documents.  The "knowledge_set" parameter must provide
52           an object of the class "AI::Categorizer::KnowledgeSet" (or a
53           subclass thereof), populated with lots of documents and categories.
54           See AI::Categorizer::KnowledgeSet for the details of how to create
55           such an object.  If you provided a "knowledge_set" parameter to
56           new(), specifying one here will override it.
57
58       categorize($document)
59           Returns an "AI::Categorizer::Hypothesis" object representing the
60           categorizer's "best guess" about which categories the given
61           document should be assigned to.  See AI::Categorizer::Hypothesis
62           for more details on how to use this object.
63
64       categorize_collection(collection => $collection)
65           Categorizes every document in a collection and returns an
66           Experiment object representing the results.  Note that the
67           Experiment does not contain knowledge of the assigned categories
68           for every document, only a statistical summary of the results.
69
70       knowledge_set()
71           Gets/sets the internal "knowledge_set" member.  Note that since the
72           knowledge set may be enormous, some Learners may throw away their
73           knowledge set after training or after restoring state from a file.
74
75       $learner->save_state($path)
76           Saves the Learner for later use.  This method is inherited from
77           "AI::Categorizer::Storable".
78
79       $class->restore_state($path)
80           Returns a Learner saved in a file with save_state().  This method
81           is inherited from "AI::Categorizer::Storable".
82

AUTHOR

84       Ken Williams, ken@mathforum.org
85
87       Copyright 2000-2003 Ken Williams.  All rights reserved.
88
89       This library is free software; you can redistribute it and/or modify it
90       under the same terms as Perl itself.
91

SEE ALSO

93       AI::Categorizer(3)
94
95
96
97perl v5.36.0                      2023-01-19       AI::Categorizer::Learner(3)
Impressum