AI::Categorizer::Learner::Boolean - Abstract class for boolean categorizers

 package AI::Categorizer::Learner::SomethingNew;
 use AI::Categorizer::Learner::Boolean;
 @ISA = qw(AI::Categorizer::Learner::Boolean);

 sub create_boolean_model {
   my ($self, $positives, $negatives, $category) = @_;
   ...
   return $something_helpful;
 }

 sub get_boolean_score {
   my ($self, $document, $something_helpful) = @_;
   ...
   return $score;
 }

This is an abstract class which turns boolean categorizers (categorizers based on algorithms that can just provide yes/no categorization decisions for a single document and single category) into multi-valued categorizers. For instance, the decision tree categorizer AI::Categorizer::Learner::DecisionTree maintains a decision tree for each category, then uses it to decide whether a certain document belongs to the given category.

Any class that inherits from this class should implement the following methods:

Used during training to create a category-specific model. The type of model you create is up to you - it should be returned as a scalar. Whatever you return will be available to you in the get_boolean_score() method, so put any information you'll need during categorization in this scalar.

In addition to $self, this method will be passed three arguments. The first argument is a reference to an array of positive examples, i.e. documents that belong to the given category. The next argument is a reference to an array of negative examples, i.e. documents that do not belong to the given category. The final argument is the Category object for the given category.

Used during categorization to assign a score for a single document relative to a single category. The score should be between 0 and 1, with a score greater than 0.5 indicating membership in the category.

In addition to $self, this method will be passed two arguments. The first argument is the document to be categorized. The second argument is the value returned by create_boolean_model() for this category.

Ken Williams, <ken@mathforum.org>

AI::Categorizer