cc.mallet.classify.tests.TestNaiveBayes Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of mallet Show documentation
MALLET is a Java-based package for statistical natural language processing, document classification, clustering, topic modeling, information extraction, and other machine learning applications to text.
The newest version!
/* Copyright (C) 2002 Univ. of Massachusetts Amherst, Computer Science Dept.
   This file is part of "MALLET" (MAchine Learning for LanguagE Toolkit).
   http://www.cs.umass.edu/~mccallum/mallet
   This software is provided under the terms of the Common Public License,
   version 1.0, as published by http://www.opensource.org.  For further
   information, see the file `LICENSE' included with this distribution. */




/** 
   @author Andrew McCallum [email protected]
 */

package cc.mallet.classify.tests;

import junit.framework.*;
import java.net.URI;
import java.io.File;

import cc.mallet.classify.*;
import cc.mallet.pipe.*;
import cc.mallet.pipe.iterator.ArrayIterator;
import cc.mallet.pipe.iterator.FileIterator;
import cc.mallet.types.*;
import cc.mallet.util.*;

public class TestNaiveBayes extends TestCase
{
	public TestNaiveBayes (String name)
	{
		super (name);
	}

	public void testNonTrained ()
	{
		Alphabet fdict = new Alphabet ();
		System.out.println ("fdict.size="+fdict.size());
		LabelAlphabet ldict = new LabelAlphabet ();
		Multinomial.Estimator me1 = new Multinomial.LaplaceEstimator (fdict);
		Multinomial.Estimator me2 = new Multinomial.LaplaceEstimator (fdict);

		// Prior
		ldict.lookupIndex ("sports");
		ldict.lookupIndex ("politics");
		ldict.stopGrowth ();
		System.out.println ("ldict.size="+ldict.size());
		Multinomial prior = new Multinomial (new double[] {.5, .5}, ldict);

		// Sports
		me1.increment ("win", 5);
		me1.increment ("puck", 5);
		me1.increment ("team", 5);
		System.out.println ("fdict.size="+fdict.size());

		// Politics
		me2.increment ("win", 5);
		me2.increment ("speech", 5);
		me2.increment ("vote", 5);

		Multinomial sports = me1.estimate();
		Multinomial politics = me2.estimate();

		// We must estimate from me1 and me2 after all data is incremented,
		// so that the "sports" multinomial knows the full dictionary size!

		Classifier c = new NaiveBayes (new Noop (fdict, ldict),
				prior,
				new Multinomial[] {sports, politics});

		Instance inst = c.getInstancePipe().instanceFrom(
				new Instance (new FeatureVector (fdict,
						new Object[] {"speech", "win"},
						new double[] {1, 1}),
						ldict.lookupLabel ("politics"),
						null, null));
		System.out.println ("inst.data = "+inst.getData ());

		Classification cf = c.classify (inst);
		LabelVector l = (LabelVector) cf.getLabeling();
		//System.out.println ("l.size="+l.size());
		System.out.println ("l.getBestIndex="+l.getBestIndex());
		assertTrue (cf.getLabeling().getBestLabel()
				== ldict.lookupLabel("politics"));
		assertTrue (cf.getLabeling().getBestValue()	> 0.6);
	}

	public void testStringTrained ()
	{
		String[] africaTraining = new String[] {
				"on the plains of africa the lions roar",
				"in swahili ngoma means to dance",
				"nelson mandela became president of south africa",
		"the saraha dessert is expanding"};
		String[] asiaTraining = new String[] {
				"panda bears eat bamboo",
				"china's one child policy has resulted in a surplus of boys",
		"tigers live in the jungle"};

		InstanceList instances =
			new InstanceList (
					new SerialPipes (new Pipe[] {
							new Target2Label (),
							new CharSequence2TokenSequence (),
							new TokenSequence2FeatureSequence (),
							new FeatureSequence2FeatureVector ()}));

		instances.addThruPipe (new ArrayIterator (africaTraining, "africa"));
		instances.addThruPipe (new ArrayIterator (asiaTraining, "asia"));
		Classifier c = new NaiveBayesTrainer ().train (instances);

		Classification cf = c.classify ("nelson mandela never eats lions");
		assertTrue (cf.getLabeling().getBestLabel()
				== ((LabelAlphabet)instances.getTargetAlphabet()).lookupLabel("africa"));
	}

	public void testRandomTrained ()
	{
		InstanceList ilist = new InstanceList (new Randoms(1), 10, 2);
		Classifier c = new NaiveBayesTrainer ().train (ilist);
		// test on the training data
		int numCorrect = 0;
		for (int i = 0; i < ilist.size(); i++) {
			Instance inst = ilist.get(i);
			Classification cf = c.classify (inst);
			cf.print ();
			if (cf.getLabeling().getBestLabel() == inst.getLabeling().getBestLabel())
				numCorrect++;
		}
		System.out.println ("Accuracy on training set = " + ((double)numCorrect)/ilist.size());
	}

	public void testIncrementallyTrainedGrowingAlphabets()
	{
		System.out.println("testIncrementallyTrainedGrowingAlphabets");
		String[]    args = new String[] {
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/a",
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/b"
		};

		File[] directories = new File[args.length];
		for (int i = 0; i < args.length; i++)
			directories[i] = new File (args[i]);

		SerialPipes instPipe =
			// MALLET pipeline for converting instances to feature vectors
			new SerialPipes(new Pipe[] {
					new Target2Label(),
					new Input2CharSequence(),
					//SKIP_HEADER only works for Unix
					//new CharSubsequence(CharSubsequence.SKIP_HEADER),
					new CharSequence2TokenSequence(),
					new TokenSequenceLowercase(),
					new TokenSequenceRemoveStopwords(),
					new TokenSequence2FeatureSequence(),
					new FeatureSequence2FeatureVector() });

		InstanceList instList = new InstanceList(instPipe);
		instList.addThruPipe(new
				FileIterator(directories, FileIterator.STARTING_DIRECTORIES));

		System.out.println("Training 1");
		NaiveBayesTrainer trainer = new NaiveBayesTrainer();
		NaiveBayes classifier = trainer.trainIncremental(instList);

		//instList.getDataAlphabet().stopGrowth();

		// incrementally train...
		String[] t2directories = {
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/b"
		};

		System.out.println("data alphabet size " + instList.getDataAlphabet().size());
		System.out.println("target alphabet size " + instList.getTargetAlphabet().size());
		InstanceList instList2 = new InstanceList(instPipe);
		instList2.addThruPipe(new
				FileIterator(t2directories, FileIterator.STARTING_DIRECTORIES));

		System.out.println("Training 2");

		System.out.println("data alphabet size " + instList2.getDataAlphabet().size());
		System.out.println("target alphabet size " + instList2.getTargetAlphabet().size());

		NaiveBayes classifier2 = (NaiveBayes) trainer.trainIncremental(instList2);
	}

	public void testIncrementallyTrained()
	{
		System.out.println("testIncrementallyTrained");
		String[]    args = new String[] {
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/a",
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/b"
		};

		File[] directories = new File[args.length];
		for (int i = 0; i < args.length; i++)
			directories[i] = new File (args[i]);

		SerialPipes instPipe =
			// MALLET pipeline for converting instances to feature vectors
			new SerialPipes(new Pipe[] {
					new Target2Label(),
					new Input2CharSequence(),
					//SKIP_HEADER only works for Unix
					//new CharSubsequence(CharSubsequence.SKIP_HEADER),
					new CharSequence2TokenSequence(),
					new TokenSequenceLowercase(),
					new TokenSequenceRemoveStopwords(),
					new TokenSequence2FeatureSequence(),
					new FeatureSequence2FeatureVector() });

		InstanceList instList = new InstanceList(instPipe);
		instList.addThruPipe(new
				FileIterator(directories, FileIterator.STARTING_DIRECTORIES));

		System.out.println("Training 1");
		NaiveBayesTrainer trainer = new NaiveBayesTrainer();
		NaiveBayes classifier = (NaiveBayes) trainer.trainIncremental(instList);

		Classification initialClassification = classifier.classify("Hello Everybody");
		Classification initial2Classification = classifier.classify("Goodbye now");
		System.out.println("Initial Classification = ");
		initialClassification.print();
		initial2Classification.print();
		System.out.println("data alphabet " + classifier.getAlphabet());
		System.out.println("label alphabet " + classifier.getLabelAlphabet());


		// incrementally train...
		String[] t2directories = {
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/b"
		};

		System.out.println("data alphabet size " + instList.getDataAlphabet().size());
		System.out.println("target alphabet size " + instList.getTargetAlphabet().size());
		InstanceList instList2 = new InstanceList(instPipe);
		instList2.addThruPipe(new
				FileIterator(t2directories, FileIterator.STARTING_DIRECTORIES));

		System.out.println("Training 2");

		System.out.println("data alphabet size " + instList2.getDataAlphabet().size());
		System.out.println("target alphabet size " + instList2.getTargetAlphabet().size());

		NaiveBayes classifier2 = (NaiveBayes) trainer.trainIncremental(instList2);


	}

	public void testEmptyStringBug()
	{
		System.out.println("testEmptyStringBug");
		String[]    args = new String[] {
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/a",
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/b"
		};

		File[] directories = new File[args.length];
		for (int i = 0; i < args.length; i++)
			directories[i] = new File (args[i]);

		SerialPipes instPipe =
			// MALLET pipeline for converting instances to feature vectors
			new SerialPipes(new Pipe[] {
					new Target2Label(),
					new Input2CharSequence(),
					//SKIP_HEADER only works for Unix
					//new CharSubsequence(CharSubsequence.SKIP_HEADER),
					new CharSequence2TokenSequence(),
					new TokenSequenceLowercase(),
					new TokenSequenceRemoveStopwords(),
					new TokenSequence2FeatureSequence(),
					new FeatureSequence2FeatureVector() });

		InstanceList instList = new InstanceList(instPipe);
		instList.addThruPipe(new
				FileIterator(directories, FileIterator.STARTING_DIRECTORIES));

		System.out.println("Training 1");
		NaiveBayesTrainer trainer = new NaiveBayesTrainer();
		NaiveBayes classifier = (NaiveBayes) trainer.trainIncremental(instList);

		Classification initialClassification = classifier.classify("Hello Everybody");
		Classification initial2Classification = classifier.classify("Goodbye now");
		System.out.println("Initial Classification = ");
		initialClassification.print();
		initial2Classification.print();
		System.out.println("data alphabet " + classifier.getAlphabet());
		System.out.println("label alphabet " + classifier.getLabelAlphabet());


		// test
		String[] t2directories = {
				"src/cc/mallet/classify/tests/NaiveBayesData/learn/b"
		};

		System.out.println("data alphabet size " + instList.getDataAlphabet().size());
		System.out.println("target alphabet size " + instList.getTargetAlphabet().size());
		InstanceList instList2 = new InstanceList(instPipe);
		instList2.addThruPipe(new
				FileIterator(t2directories, FileIterator.STARTING_DIRECTORIES, true));

		System.out.println("Training 2");

		System.out.println("data alphabet size " + instList2.getDataAlphabet().size());
		System.out.println("target alphabet size " + instList2.getTargetAlphabet().size());

		NaiveBayes classifier2 = (NaiveBayes) trainer.trainIncremental(instList2);
		Classification secondClassification = classifier.classify("Goodbye now");
		secondClassification.print();

	}




	static Test suite ()
	{
		return new TestSuite (TestNaiveBayes.class);
		//TestSuite suite= new TestSuite();
		//   //suite.addTest(new TestNaiveBayes("testIncrementallyTrained"));
		// suite.addTest(new TestNaiveBayes("testEmptyStringBug"));

		// return suite;
	}

	protected void setUp ()
	{
	}

	public static void main (String[] args)
	{
		junit.textui.TestRunner.run (suite());
	}

}