Package opennlp.tools.ml.naivebayes
Class NaiveBayesTrainer
java.lang.Object
opennlp.tools.ml.AbstractTrainer
opennlp.tools.ml.AbstractEventTrainer
opennlp.tools.ml.naivebayes.NaiveBayesTrainer
- All Implemented Interfaces:
Trainer,EventTrainer
Trains
models using the combination of EM algorithm
and Naive Bayes classifier which is described in:
Text Classification from Labeled and Unlabeled Documents using EM Nigam, McCallum, et al. paper of 2000
- See Also:
-
Field Summary
FieldsFields inherited from class opennlp.tools.ml.AbstractEventTrainer
DATA_INDEXER_ONE_PASS_REAL_VALUE, DATA_INDEXER_ONE_PASS_VALUE, DATA_INDEXER_PARAM, DATA_INDEXER_TWO_PASS_VALUEFields inherited from class opennlp.tools.ml.AbstractTrainer
ALGORITHM_PARAM, CUTOFF_DEFAULT, CUTOFF_PARAM, ITERATIONS_DEFAULT, ITERATIONS_PARAM, TRAINER_TYPE_PARAMFields inherited from interface opennlp.tools.ml.EventTrainer
EVENT_VALUE -
Constructor Summary
ConstructorsConstructorDescriptionInstantiates aNaiveBayesTrainerwith default training parameters.NaiveBayesTrainer(TrainingParameters parameters) Instantiates aNaiveBayesTrainerwith specificTrainingParameters. -
Method Summary
Modifier and TypeMethodDescriptiondoTrain(DataIndexer indexer) booleanTrains aNaiveBayesModelwith given parameters.Methods inherited from class opennlp.tools.ml.AbstractEventTrainer
getDataIndexer, train, train, validateMethods inherited from class opennlp.tools.ml.AbstractTrainer
getAlgorithm, getCutoff, getIterations, init
-
Field Details
-
NAIVE_BAYES_VALUE
- See Also:
-
-
Constructor Details
-
NaiveBayesTrainer
public NaiveBayesTrainer()Instantiates aNaiveBayesTrainerwith default training parameters. -
NaiveBayesTrainer
Instantiates aNaiveBayesTrainerwith specificTrainingParameters.- Parameters:
parameters- Theparameterto use.
-
-
Method Details
-
isSortAndMerge
public boolean isSortAndMerge()- Specified by:
isSortAndMergein classAbstractEventTrainer
-
doTrain
- Specified by:
doTrainin classAbstractEventTrainer- Throws:
IOException
-
trainModel
Trains aNaiveBayesModelwith given parameters.- Parameters:
di- TheDataIndexerused as data input.- Returns:
- A valid, trained
Naive Bayes model.
-