weka.classifiers.bayes.NaiveBayesMultinomialUpdateable Maven / Gradle / Ivy
Show all versions of weka-dev Show documentation
/*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see
*
* Andrew Mccallum, Kamal Nigam: A Comparison of Event Models for Naive Bayes Text Classification. In: AAAI-98 Workshop on 'Learning for Text Categorization', 1998.
*
* The core equation for this classifier:
*
* P[Ci|D] = (P[D|Ci] x P[Ci]) / P[D] (Bayes rule)
*
* where Ci is class i and D is a document.
*
*
* BibTeX:
*
* @inproceedings{Mccallum1998,
* author = {Andrew Mccallum and Kamal Nigam},
* booktitle = {AAAI-98 Workshop on 'Learning for Text Categorization'},
* title = {A Comparison of Event Models for Naive Bayes Text Classification},
* year = {1998}
* }
*
*
*
* Valid options are:
*
* -output-debug-info
* If set, classifier is run in debug mode and may output additional info to
* the console.
*
*
* -do-not-check-capabilities
* If set, classifier capabilities are not checked before classifier is built
* (use with caution).
*
*
* -num-decimal-laces
* The number of decimal places for the output of numbers in the model.
*
*
* -batch-size
* The desired batch size for batch prediction.
*
*
*
* @author Andrew Golightly ([email protected])
* @author Bernhard Pfahringer ([email protected])
* @author Eibe Frank ([email protected])
* @version $Revision: 14250 $
*/
public class NaiveBayesMultinomialUpdateable extends NaiveBayesMultinomial implements UpdateableClassifier {
/** for serialization */
static final long serialVersionUID = -7204398796974263186L;
/** the number of words per class. */
protected double[] m_wordsPerClass;
/**
* Returns a string describing this classifier
* @return a description of the classifier suitable for
* displaying in the explorer/experimenter gui
*/
public String globalInfo() {
return
"Class for building and using an updateable multinomial Naive Bayes classifier. "
+ "For more information see,\n\n"
+ getTechnicalInformation().toString() + "\n\n"
+ "The core equation for this classifier:\n\n"
+ "P[Ci|D] = (P[D|Ci] x P[Ci]) / P[D] (Bayes' rule)\n\n"
+ "where Ci is class i and D is a document.";
}
/**
* Generates the classifier.
*
* @param instances set of instances serving as training data
* @throws Exception if the classifier has not been generated successfully
*/
public void buildClassifier(Instances instances) throws Exception {
initializeClassifier(instances);
//enumerate through the instances
m_wordsPerClass = new double[m_numClasses];
for (int i = 0; i < m_numClasses; i++) {
m_wordsPerClass[i] = m_numAttributes - 1;
}
for (Instance instance : instances) {
updateClassifier(instance);
}
}
/**
* Updates the classifier with information from one training instance.
*
* @param instance the instance to be incorporated
* @throws Exception if the instance cannot be processed successfully.
*/
public void updateClassifier(Instance instance) throws Exception {
double classValue = instance.value(instance.classIndex());
if (!Utils.isMissingValue(classValue)) {
int classIndex = (int) classValue;
m_probOfClass[classIndex] += instance.weight();
for (int a = 0; a < instance.numValues(); a++) {
if (instance.index(a) != instance.classIndex()) {
if (!instance.isMissingSparse(a)) {
double numOccurrences = instance.valueSparse(a) * instance.weight();
if (numOccurrences < 0)
throw new Exception("Numeric attribute values must all be greater or equal to zero.");
m_wordsPerClass[classIndex] += numOccurrences;
m_probOfWordGivenClass[classIndex][instance.index(a)] += numOccurrences;
}
}
}
}
}
/**
* log(N!) + (sum for all the words i)(log(Pi^ni) - log(ni!))
*
* where
* N is the total number of words
* Pi is the probability of obtaining word i
* ni is the number of times the word at index i occurs in the document
*
* Actually, this method just computes (sum for all the words i)(log(Pi^ni) because the factorials are irrelevant
* when posterior class probabilities are computed.
*
* @param inst The instance to be classified
* @param classIndex The index of the class we are calculating the probability with respect to
*
* @return The log of the probability of the document occuring given the class
*/
protected double probOfDocGivenClass(Instance inst, int classIndex) {
double answer = 0;
for(int i = 0; i < inst.numValues(); i++) {
if (inst.index(i) != inst.classIndex()) {
answer += inst.valueSparse(i) * (Math.log(m_probOfWordGivenClass[classIndex][inst.index(i)]) -
Math.log(m_wordsPerClass[classIndex]));
}
}
return answer;
}
/**
* Returns a string representation of the classifier.
*
* @return a string representation of the classifier
*/
public String toString()
{
StringBuffer result = new StringBuffer("The class counts (including Laplace correction)\n-----------------------------------------------\n");
for(int c = 0; c