Examples of statechum.analysis.learning.MarkovModel

statechum.analysis.learning.MarkovModel
Describes a non-probabilistic Markov model, where for every path we know either that,
- the path was never encountered or
- the path was encountered and there is a specific set of elements of alphabet that followed it.
The idea is to use the supplied Markov matrix to predict transitions from a specific state, passed as an argument. The choice of direction is not a choice between predicting transitions leaving a state based on those surrounding that state v.s predicting transitions entering a state based on those surrounding it. It is rather a choice of classifier to make predictions, the one that looks at history and decides what is to follow and the one looking at surrounding transitions and making decisions based on that.
- Where predictForwardOrSideways is true, we are predicting transitions based on paths leading to the state of interest. Parameter Inverse_Graph should be the (non-deterministic) inverse of graph.
- Where predictForwardOrSideways is false, we are predicting transitions based on paths leading from the state of interest (sideways predictions). Parameter Inverse_Graph should be the same as graph and pathBeyondCurrentState should be null because once we predicted one transition, there are no further transitions from that state, hence no further transitions can be predicted sideways.

  }
  
  @Test
  public void testCreateMarkovMatrix4()
  {
    MarkovModel m = new MarkovModel(2,true,true);
    Set<List<Label>> plusStrings = new HashSet<List<Label>>(), minusStrings = buildSet(new String[][] { new String[]{"u"} },config,converter);
    Map<Trace, MarkovOutcome> matrix = m.createMarkovLearner(plusStrings, minusStrings,false);
    Assert.assertEquals(1,matrix.size());


    Assert.assertSame(MarkovOutcome.negative, matrix.get(new Trace(Arrays.asList(new Label[]{lblU}),true)));
  }

View Full Code Here

  
  /** Same as testCreateMarkovMatrix3 but contains an empty trace which is ignored since it does not match any of the valid chunk sizes. */
  @Test
  public void testCreateMarkovMatrix5()
  {
    final MarkovModel m = new MarkovModel(2,true,true);
    final Set<List<Label>> plusStrings = new HashSet<List<Label>>(), minusStrings = buildSet(new String[][] { new String[]{},new String[]{"a","u"} },config,converter);
    Map<Trace, MarkovOutcome> matrix = m.createMarkovLearner(plusStrings, minusStrings,false);
    Assert.assertEquals(3,matrix.size());
    
    Assert.assertSame(MarkovOutcome.negative, matrix.get(new Trace(Arrays.asList(new Label[]{lblA,lblU}),true)));


    Assert.assertSame(MarkovOutcome.positive, matrix.get(new Trace(Arrays.asList(new Label[]{lblA}),true)));

View Full Code Here

  }
  
  @Test
  public void testCreateMarkovMatrix6()
  {
    final MarkovModel m = new MarkovModel(2,true,true);
    final Set<List<Label>> plusStrings = new HashSet<List<Label>>(), minusStrings = new HashSet<List<Label>>();
    Helper.checkForCorrectException(new whatToRun() {
      @Override
      public void run() throws NumberFormatException
      {
        m.createMarkovLearner(plusStrings, minusStrings,false);
      }
    }, IllegalArgumentException.class, "empty");
  }

View Full Code Here

  }
  
  @Test
  public void testCreateMarkovMatrix7()
  {
    final MarkovModel m = new MarkovModel(2,true,true);
    final Set<List<Label>> plusStrings = new HashSet<List<Label>>(), minusStrings = buildSet(new String[][] { new String[]{},new String[]{} },config,converter);
    Helper.checkForCorrectException(new whatToRun() {
      @Override
      public void run() throws NumberFormatException
      {
        m.createMarkovLearner(plusStrings, minusStrings,false);
      }
    }, IllegalArgumentException.class, "empty");
  }

View Full Code Here

  
  /** Nothing to add because there not enough evidence. */
  @Test
  public void testConstructExtendedGraph1()
  {
    MarkovModel m = new MarkovModel(2,true,true);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","p"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-u->B-p->B","testConstructExtendedGraph1",config, converter);
    MarkovClassifier cl = new MarkovClassifier(m,graph);
    Map<CmpVertex, Map<Label, MarkovOutcome>> newTransitions = cl.predictTransitions();
    Assert.assertTrue(newTransitions.isEmpty());// not enough evidence to update, hence nothing should be recorded.
    final LearnerGraph expected = FsmParser.buildLearnerGraph("A-u->B-p->B","testConstructExtendedGraph1",config, converter);

View Full Code Here

  
  /** Nothing to add because the alphabet of the graph of interest consists of a single letter for which there is no statistical data. */
  @Test
  public void testConstructExtendedGraph2()
  {
    MarkovModel m = new MarkovModel(2,true,true);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","p"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B","testConstructExtendedGraph2",config, converter);
    MarkovClassifier cl = new MarkovClassifier(m,graph);
    Map<CmpVertex, Map<Label, MarkovOutcome>> newTransitions = cl.predictTransitions();
    Assert.assertTrue(newTransitions.isEmpty());// not enough evidence to update, hence nothing should be recorded.
    final LearnerGraph expected = FsmParser.buildLearnerGraph("A-a->B","testConstructExtendedGraph2",config, converter);

View Full Code Here

  
  /** A bit has been added. */
  @Test
  public void testConstructExtendedGraph3()
  {
    MarkovModel m = new MarkovModel(2,true,true);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B / T-b->T-u->T","testConstructExtendedGraph3a",config, converter);
    MarkovClassifier cl = new MarkovClassifier(m,graph);
    Map<CmpVertex, Map<Label, MarkovOutcome>> newTransitions = cl.predictTransitions();
    Assert.assertEquals(1,newTransitions.size());// not enough evidence to update, hence nothing should be recorded.

View Full Code Here

  
  /** A bit has been added, but reject-transition was not because is present both in the positive and negative light in the initial traces. */
  @Test
  public void testConstructExtendedGraph4()
  {
    MarkovModel m = new MarkovModel(2,true,true);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"a","u"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B / T-b->T-u->T","testConstructExtendedGraph3a",config, converter);
    MarkovClassifier cl = new MarkovClassifier(m,graph);
    Map<CmpVertex, Map<Label, MarkovOutcome>> newTransitions = cl.predictTransitions();
    Assert.assertEquals(1,newTransitions.size());// not enough evidence to update, hence nothing should be recorded.

View Full Code Here

  
  /** A bit has been added, but reject-transition was not because is present both in the positive and negative light in the initial traces. */
  @Test
  public void testConstructExtendedGraph5()
  {
    MarkovModel m = new MarkovModel(2,true,true);// w below is to ensure that all elements of the alphabet are included in traces.
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","u"},new String[]{"w"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B / A-w->M-c->B / T-b->T-u->T","testConstructExtendedGraph5a",config, converter);// the purpose of the w-transition is to ensure transition c is taken into account in graph comparison
    MarkovClassifier cl = new MarkovClassifier(m,graph);
    Map<CmpVertex, Map<Label, MarkovOutcome>> newTransitions = cl.predictTransitions();
    Assert.assertEquals(1,newTransitions.size());

View Full Code Here

  
  /** Nothing has been added. u has been seen both in the positive and negative light and lead to inconsistency. */
  @Test
  public void testConstructExtendedGraph6()
  {
    MarkovModel m = new MarkovModel(2,true,true);
    Set<List<Label>> plusStrings = buildSet(new String[][] { new String[]{"a","b"},new String[]{"c","u"} },config,converter), minusStrings = buildSet(new String[][] { new String[]{"a","u"} },config,converter);
    m.createMarkovLearner(plusStrings, minusStrings,false);
    final LearnerGraph graph = FsmParser.buildLearnerGraph("A-a->B / A-c->B / T-b->T-u->T","testConstructExtendedGraph6a",config, converter);
    MarkovClassifier cl = new MarkovClassifier(m,graph);
    Map<CmpVertex, Map<Label, MarkovOutcome>> newTransitions = cl.predictTransitions();
    
    Assert.assertEquals(1,newTransitions.size());

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of statechum.analysis.learning.MarkovModel

statechum.analysis.Erlang.Synapse$StatechumProcess

statechum.analysis.learning.experiments.PairSelection.ASE2014$LearnerRunner

statechum.analysis.learning.experiments.PairSelection.Cav2014$EvaluationOfExisingLearnerRunner

statechum.analysis.learning.experiments.PairSelection.Cav2014$LearnerRunner

statechum.analysis.learning.experiments.PairSelection.CVS$LearnerRunner

statechum.analysis.learning.experiments.PairSelection.CVS_With_Random_traces_Generation$LearnerRunner

statechum.analysis.learning.experiments.PairSelection.LearnFromTracesUsingExperimentalLearners

statechum.analysis.learning.experiments.PairSelection.MarkovLearnerExperimentWithStatisticalAnalysis$LearnerRunner

statechum.analysis.learning.experiments.PairSelection.MarkovPassivePairSelection

statechum.analysis.learning.experiments.PairSelection.MarkovPassivePairSelection$LearnerRunner

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.