1AI::Categorizer::LearneUrs:e:rSVCMo(n3t)ributed Perl DocAuIm:e:nCtaatteigoonrizer::Learner::SVM(3)
2
3
4

NAME

6       AI::Categorizer::Learner::SVM - Support Vector Machine Learner
7

SYNOPSIS

9         use AI::Categorizer::Learner::SVM;
10
11         # Here $k is an AI::Categorizer::KnowledgeSet object
12
13         my $l = new AI::Categorizer::Learner::SVM(...parameters...);
14         $l->train(knowledge_set => $k);
15         $l->save_state('filename');
16
17         ... time passes ...
18
19         $l = AI::Categorizer::Learner->restore_state('filename');
20         while (my $document = ... ) {  # An AI::Categorizer::Document object
21           my $hypothesis = $l->categorize($document);
22           print "Best assigned category: ", $hypothesis->best_category, "\n";
23         }
24

DESCRIPTION

26       This class implements a Support Vector Machine machine learner, using
27       Cory Spencer's "Algorithm::SVM" module.  In lots of the recent academic
28       literature, SVMs perform very well for text categorization.
29

METHODS

31       This class inherits from the "AI::Categorizer::Learner" class, so all
32       of its methods are available unless explicitly mentioned here.
33
34   new()
35       Creates a new SVM Learner and returns it.  In addition to the
36       parameters accepted by the "AI::Categorizer::Learner" class, the SVM
37       subclass accepts the following parameters:
38
39       svm_kernel
40           Specifies what type of kernel should be used when building the SVM.
41           Default is 'linear'.  Possible values are 'linear', 'polynomial',
42           'radial' and 'sigmoid'.
43
44   train(knowledge_set => $k)
45       Trains the categorizer.  This prepares it for later use in categorizing
46       documents.  The "knowledge_set" parameter must provide an object of the
47       class "AI::Categorizer::KnowledgeSet" (or a subclass thereof),
48       populated with lots of documents and categories.  See
49       AI::Categorizer::KnowledgeSet for the details of how to create such an
50       object.
51
52   categorize($document)
53       Returns an "AI::Categorizer::Hypothesis" object representing the
54       categorizer's "best guess" about which categories the given document
55       should be assigned to.  See AI::Categorizer::Hypothesis for more
56       details on how to use this object.
57
58   save_state($path)
59       Saves the categorizer for later use.  This method is inherited from
60       "AI::Categorizer::Storable".
61

AUTHOR

63       Ken Williams, ken@mathforum.org
64

66       Copyright 2000-2003 Ken Williams.  All rights reserved.
67
68       This library is free software; you can redistribute it and/or modify it
69       under the same terms as Perl itself.
70

SEE ALSO

72       AI::Categorizer(3), Algorithm::SVM(3)
73
74
75
76perl v5.32.1                      2021-01-26  AI::Categorizer::Learner::SVM(3)
Impressum