PRTools contents

PRTools manual

naivebcc

NAIVEBCC

Naive Bayes Combining Classifier

    W = A*(WU*NAIVEBCC)
    W = WT*NAIVEBCC(B*WT)
    D = C*W

Input
 A Dataset used for training base classifiers as well as combiner
 B Dataset used for training combiner of trained base classifiers
 C Dataset used for testing (executing) the combiner
 WU Set of untrained base classifiers, see STACKED
 WT Set of trained base classifiers, see STACKED

Output
 W Trained Naive Bayes Combining Classifier
 D Dataset with prob. products (over base classifiers) per class

Description

During training the combiner computes the probabilities  P(class | classifier outcomes) based on the crisp class assignements  made by the base classifiers for the training set. During execution the  product of these probabilities are computed, again following the crisp  class assignments of the base classifiers. These products are returned  as columns in D. Use CLASSC to normalise the outcomes. Use TESTD or  LABELD to inspect performances and assigned labels.

NAIVEBCC differs from the classifier NAIVEBC by the fact that the  latter uses continuous inputs (no crisp labeling) and does not make a  distinction between classifiers. Like almost any other classifier  however, NAIVEBC may be used as a trained combiner as well.

Reference(s)

1. Kuncheva, LI. Combining pattern classifiers, 2004, pp.126-128.

See also

datasets, mappings, stacked, naivebc, classc, testd, labeld,

PRTools contents

PRTools manual