import com.aliasi.classify.Classification; //导入依赖的package包/类
protected void train(ReviewCorpus trainerCorpus)
        throws FileNotFoundException, IOException {
    HashMap<String, List<String>> categoriesWithSentences = getCategoriesWithSentences(
            trainerCorpus);
    int numberOfCategories = 0;
    int numberOfSentences = 0;
    for (String category : classifier.categories()) {
        logger.debug("Training: {}", category);
        Classification classification = new Classification(category);
        numberOfCategories++;
        numberOfSentences = 0;
        for (String sentence : categoriesWithSentences.get(category)) {
            logger.debug("\t sentence: {}", sentence);
            Classified<CharSequence> classified = new Classified<CharSequence>(
                    sentence, classification);
            classifier.handle(classified);
            numberOfSentences++;
        }
        logger.debug("Training of {} finished. Number of sentences used: {}",
                category, numberOfSentences);
    }
    logger.debug("Training finished. Number of categories: {}.",
            numberOfCategories);
}
 

import com.aliasi.classify.Classification; //导入依赖的package包/类
public void evaluate(ReviewCorpus evaluatorCorpus) {
    JointClassifierEvaluator<CharSequence> evaluator = new JointClassifierEvaluator<CharSequence>(
            classifier, classifier.categories(), false);
    HashMap<String, List<String>> categoriesWithSentences = getCategoriesWithSentences(
            evaluatorCorpus);
    int numberOfCategories = 0;
    int numberOfSentences = 0;
    for (String category : getCategories(evaluatorCorpus)) {
        Classification classification = new Classification(category);
        numberOfCategories++;
        numberOfSentences = 0;
        for (String sentence : categoriesWithSentences.get(category)) {
            Classified<CharSequence> classified = new Classified<CharSequence>(
                    sentence, classification);
            evaluator.handle(classified);
            numberOfSentences++;
        }
        logger.debug("Evaluation of {} finished. Number of sentences used: {}",
                category, numberOfSentences);
    }
    logger.debug(
            "Evaluation of {} finished. Number of categories: {}. Details: {}",
            this.getClass(), numberOfCategories, evaluator);
}
 

import com.aliasi.classify.Classification; //导入依赖的package包/类
public void buildSentimentAnalysisModel() {
        out.println("Building Sentiment Model");
        
        File trainingDir = new File("C:\\Jenn Personal\\Packt Data Science\\Chapter 12\\review_polarity\\txt_sentoken");
        for (int i = 0; i < labels.length; i++) {
            Classification classification = new Classification(labels[i]);
            File file = new File(trainingDir, labels[i]);
            File[] trainingFiles = file.listFiles();
            for (int j = 0; j < trainingFiles.length; j++) {
                try {
                    String review = Files.readFromFile(trainingFiles[j], "ISO-8859-1");
                    Classified<CharSequence> classified = new Classified<>(review, classification);
                    classifier.handle(classified);
                } catch (IOException ex) {
                    ex.printStackTrace();
                }
            }
        }

//        out.println("---saving model");
//        try {
//            AbstractExternalizable
//                    .compileTo((Compilable) classifier, 
//                            new File("classifier.model"));
//            out.println("---classifer model saved");
//        } catch (IOException ex) {
//            ex.printStackTrace();
//        }
//        out.println("---buildSentimentAnalysisModel terminated");
    }
 

import com.aliasi.classify.Classification; //导入依赖的package包/类
private void trainClassifier(Set<String> categorySet, List<String[]> annotatedData){
    String[] categories = categorySet.toArray(new String[0]);
     classifier = DynamicLMClassifier.createNGramBoundary(categories,maxCharNGram);
    for (String[] row: annotatedData) {
        String actualClassification = row[0];
        String text = row[1];
        Classification classification = new Classification(actualClassification);
        Classified<CharSequence> classified = new Classified<>(text,classification);
        classifier.handle(classified);
    }
}
 

import com.aliasi.classify.Classification; //导入依赖的package包/类
@Override
public void visitTrain(ObjectHandler<Classified<CharSequence>> classifier) {
	for (int i = 0; i < N; i++) {
		Classification label = new Classification(y.get(i));
		Classified<CharSequence> labeledInstance = 
				new Classified<CharSequence>(X.get(i), label);
		classifier.handle(labeledInstance);
	}
}
 

import com.aliasi.classify.Classification; //导入依赖的package包/类
@Override
public synchronized void train(String category, List<String> reviews) {
    Classification classification = new Classification(category);
    for (String r : reviews) {
        Classified<CharSequence> classified = new Classified<CharSequence>(r, classification);
        classifier.handle(classified);
    }
    trainingSize += reviews.size();
    isTrained = true;
}
 

import com.aliasi.classify.Classification; //导入依赖的package包/类
@Override
public String bestMatch(String text) throws MpcaClassifierNotTrainedException {
    if(!isTrained()) {
        throw new MpcaClassifierNotTrainedException();
    }
    Classification classification = classifier.classify(text);
    return classification.bestCategory();
}
 

import com.aliasi.classify.Classification; //导入依赖的package包/类
public TweetHandler performSentimentAnalysis() {
    Classification classification = classifier.classify(this.text);
    String bestCategory = classification.bestCategory();
    this.category = bestCategory;
    return this;
}