Java StanfordPosTagger类代码示例

import de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordPosTagger; //导入依赖的package包/类
public static void main(String[] args) throws UIMAException, IOException {

		// read text documents
		CollectionReaderDescription reader = CollectionReaderFactory.createReaderDescription(TextReader.class,
				TextReader.PARAM_SOURCE_LOCATION, textFolder, TextReader.PARAM_PATTERNS, textPattern,
				TextReader.PARAM_LANGUAGE, "en");

		// preprocess documents
		String[] quoteBegin = { "“", "‘" };
		List<String> quoteBeginList = Arrays.asList(quoteBegin);
		String[] quoteEnd = { "”", "’" };
		List<String> quoteEndList = Arrays.asList(quoteEnd);

		AnalysisEngineDescription segmenter = AnalysisEngineFactory.createEngineDescription(StanfordSegmenter.class);
		AnalysisEngineDescription pos = AnalysisEngineFactory.createEngineDescription(StanfordPosTagger.class,
				StanfordPosTagger.PARAM_QUOTE_BEGIN, quoteBeginList, StanfordPosTagger.PARAM_QUOTE_END, quoteEndList);
		AnalysisEngineDescription lemmatizer = AnalysisEngineFactory.createEngineDescription(StanfordLemmatizer.class);
		AnalysisEngineDescription stemmer = AnalysisEngineFactory.createEngineDescription(SnowballStemmer.class,
				SnowballStemmer.PARAM_LOWER_CASE, true);
		AnalysisEngineDescription parser = AnalysisEngineFactory.createEngineDescription(StanfordParser.class,
				StanfordParser.PARAM_MODEL_LOCATION, "lib/englishRNN.ser", StanfordParser.PARAM_MODE,
				DependenciesMode.CC_PROPAGATED, StanfordPosTagger.PARAM_QUOTE_BEGIN, quoteBeginList,
				StanfordPosTagger.PARAM_QUOTE_END, quoteEndList);

		// write annotated data to file
		AnalysisEngineDescription writer = AnalysisEngineFactory.createEngineDescription(BinaryCasWriter.class,
				BinaryCasWriter.PARAM_TARGET_LOCATION, textFolder, BinaryCasWriter.PARAM_STRIP_EXTENSION, false,
				BinaryCasWriter.PARAM_FILENAME_EXTENSION, ".bin6", BinaryCasWriter.PARAM_OVERWRITE, true);

		// print statistics
		AnalysisEngineDescription stat = AnalysisEngineFactory.createEngineDescription(CorpusStatWriter.class);

		// run pipeline
		SimplePipeline.runPipeline(reader, segmenter, pos, lemmatizer, stemmer, parser, writer, stat);
	}
 

import de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordPosTagger; //导入依赖的package包/类
public void run()
    throws Exception
{
    String language = "en";
    String corpora = "src/main/resources/raw/";
    String fileSuffix = "*.txt";

    CollectionReaderDescription trainReader = CollectionReaderFactory.createReaderDescription(
            LineTokenTagReader.class, LineTokenTagReader.PARAM_LANGUAGE, language,
            LineTokenTagReader.PARAM_SOURCE_LOCATION, corpora,
            LineTokenTagReader.PARAM_PATTERNS, fileSuffix);

    FlexTagTrainSaveModel flex = new FlexTagTrainSaveModel(trainReader,
            new File(System.getProperty("user.home") + "/Desktop/flexOut"));

    if (System.getProperty("DKPRO_HOME") == null) {
        flex.setDKProHomeFolder(System.getProperty("user.home") + "/Desktop/");
    }
    flex.setExperimentName("FlexTest");

    flex.setFeatures(TcFeatureFactory.create(LuceneCharacterNGram.class,
            LuceneCharacterNGram.PARAM_NGRAM_MIN_N, 2, LuceneCharacterNGram.PARAM_NGRAM_MAX_N,
            4, LuceneCharacterNGram.PARAM_NGRAM_USE_TOP_K, 50));

    flex.setPreprocessing(
            AnalysisEngineFactory.createEngineDescription(BreakIteratorSegmenter.class),
            AnalysisEngineFactory.createEngineDescription(StanfordPosTagger.class,  
                    StanfordPosTagger.PARAM_LANGUAGE, "en"));

    flex.execute();
}
 

import de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordPosTagger; //导入依赖的package包/类
@Test
public void testSegmentation() throws ResourceInitializationException, AnalysisEngineProcessException {

	SimplePipeline.runPipeline(jcas, desc, AnalysisEngineFactory.createEngineDescription(StanfordPosTagger.class));

	assertTrue(JCasUtil.exists(jcas, POS.class));
	assertEquals("Liebe", JCasUtil.selectByIndex(jcas, NN.class, 0).getCoveredText());
}
 

import de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordPosTagger; //导入依赖的package包/类
public static void main(String[] args) throws ResourceInitializationException, UIMAException, IOException {
	System.setProperty("java.util.logging.config.file", "src/main/resources/logging.properties");

	CollectionReaderDescription crd = CollectionReaderFactory.createReaderDescription(TextgridTEIUrlReader.class,
			TextgridTEIUrlReader.PARAM_INPUT, "src/main/resources");

	SimplePipeline.runPipeline(crd,
			/*
			 * Do segmentation.
			 */
			D.getWrappedSegmenterDescription(LanguageToolSegmenter.class),
			createEngineDescription(FigureReferenceAnnotator.class),
			createEngineDescription(SpeakerIdentifier.class, SpeakerIdentifier.PARAM_CREATE_SPEAKER_FIGURE, true),
			/*
			 * standard NLP components. This works because dkpro only sees
			 * tokens and sentences. The segmenter creates those only for
			 * the figure speech (and not for stage directions)
			 */
			createEngineDescription(StanfordPosTagger.class),
			createEngineDescription(StanfordNamedEntityRecognizer.class),
			createEngineDescription(FigureMentionDetection.class),
			/*
			 * Extract copresence network
			 */
			createEngineDescription(NetworkExtractor.class),
			/*
			 * extract mention network
			 */
			createEngineDescription(NetworkExtractor.class, NetworkExtractor.PARAM_VIEW_NAME, "MentionNetwork",
					NetworkExtractor.PARAM_NETWORK_TYPE, "MentionNetwork"),
			/*
			 * print xmi
			 */
			createEngineDescription(XmiWriter.class, XmiWriter.PARAM_TARGET_LOCATION, "target/xmi/"));
}
 

import de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordPosTagger; //导入依赖的package包/类
public static void main(String[] args) throws Exception {

		CollectionReaderDescription reader = createReaderDescription(PythagorasCorpusReader.class,
				PythagorasCorpusReader.PARAM_SOURCE_LOCATION, RESOURCES_PATH,
				PythagorasCorpusReader.PARAM_LANGUAGE, "de",
				PythagorasCorpusReader.PARAM_PATTERNS, PythagorasCorpusReader.INCLUDE_PREFIX + "**/*.docx");

		AnalysisEngineDescription tfIdfConsumer = createEngineDescription(
				createEngineDescription(BreakIteratorSegmenter.class),
				createEngineDescription(StanfordPosTagger.class,
					StanfordPosTagger.PARAM_LANGUAGE, "de"),
				createEngineDescription(StanfordLemmatizer.class),
				createEngineDescription(TfidfConsumer.class,
								TfidfConsumer.PARAM_FEATURE_PATH, Lemma.class,
								TfidfConsumer.PARAM_TARGET_LOCATION, OUTPUT_PATH,
								TfidfConsumer.PARAM_LOWERCASE, true));

		SimplePipeline.runPipeline(reader, tfIdfConsumer);
		
		//The following output is a test to check whether the consumer has built a correct model.
		DfModel dfModel = TfidfUtils.getDfModel(OUTPUT_PATH);
		System.out.println("Testing newly created model..." + "\n" + 
				"Very probable token: " + dfModel.getDf("ist") + "\n" + 
				"Impossible token: " + dfModel.getDf("impossibleToken") + "\n" +
				"Num of Documents: " + dfModel.getDocumentCount() + "\n" + 
				"Feature Path: " + dfModel.getFeaturePath() + "\n" + 
				"Lowercase?: " + dfModel.getLowercase() 
				);
        
	}
 

import de.tudarmstadt.ukp.dkpro.core.stanfordnlp.StanfordPosTagger; //导入依赖的package包/类
public static void main(String[] args) throws UIMAException, IOException {

		Logger.getRootLogger().setLevel(Level.INFO);

		// 0) parameter
		if (args.length > 0)
			textFolder = args[0];

		// 1) read text documents
		CollectionReaderDescription reader = CollectionReaderFactory.createReaderDescription(TextReader.class,
				TextReader.PARAM_SOURCE_LOCATION, textFolder, TextReader.PARAM_PATTERNS, textPattern,
				TextReader.PARAM_LANGUAGE, "en");

		// 2) process documents

		String[] quoteBegin = { "“", "‘" };
		List<String> quoteBeginList = Arrays.asList(quoteBegin);
		String[] quoteEnd = { "”", "’" };
		List<String> quoteEndList = Arrays.asList(quoteEnd);

		// tokenization and sentence splitting
		AnalysisEngineDescription segmenter = AnalysisEngineFactory.createEngineDescription(StanfordSegmenter.class,
				StanfordSegmenter.PARAM_NEWLINE_IS_SENTENCE_BREAK, "ALWAYS");

		// part-of-speech tagging
		AnalysisEngineDescription pos = AnalysisEngineFactory.createEngineDescription(StanfordPosTagger.class,
				StanfordPosTagger.PARAM_QUOTE_BEGIN, quoteBeginList, StanfordPosTagger.PARAM_QUOTE_END, quoteEndList);

		// lemmatizing
		AnalysisEngineDescription lemmatizer = AnalysisEngineFactory.createEngineDescription(StanfordLemmatizer.class,
				StanfordLemmatizer.PARAM_QUOTE_BEGIN, quoteBeginList, StanfordLemmatizer.PARAM_QUOTE_END, quoteEndList);

		// named entity recognition
		AnalysisEngineDescription ner = AnalysisEngineFactory.createEngineDescription(
				StanfordNamedEntityRecognizer.class, StanfordNamedEntityRecognizer.PARAM_QUOTE_BEGIN, quoteBeginList,
				StanfordNamedEntityRecognizer.PARAM_QUOTE_END, quoteEndList);

		// constituency parsing and dependency conversion
		AnalysisEngineDescription parser = AnalysisEngineFactory.createEngineDescription(StanfordParser.class,
				StanfordParser.PARAM_QUOTE_BEGIN, quoteBeginList, StanfordParser.PARAM_QUOTE_END, quoteEndList,
				StanfordParser.PARAM_MODE, DependenciesMode.CC_PROPAGATED);

		// coreference resolution
		AnalysisEngineDescription coref = AnalysisEngineFactory.createEngineDescription();

		// 3) write annotated data to file
		AnalysisEngineDescription writer = AnalysisEngineFactory.createEngineDescription(BinaryCasWriter.class,
				BinaryCasWriter.PARAM_TARGET_LOCATION, textFolder, BinaryCasWriter.PARAM_STRIP_EXTENSION, false,
				BinaryCasWriter.PARAM_FILENAME_EXTENSION, ".bin6", BinaryCasWriter.PARAM_OVERWRITE, true);

		// print statistics
		AnalysisEngineDescription stat = AnalysisEngineFactory.createEngineDescription(CorpusStatWriter.class);

		// 4) run pipeline
		SimplePipeline.runPipeline(reader, segmenter, pos, lemmatizer, ner, parser, coref, writer, stat);
	}