Example usage for org.apache.mahout.classifier ClassifierResult getScore

Introduction

In this page you can find the example usage for org.apache.mahout.classifier ClassifierResult getScore.

Prototype

public double getScore() 

Source Link

Usage

From source file:com.tamingtext.classifier.bayes.ClassifyDocument.java

License:Apache License

public static void main(String[] args) {
    log.info("Command-line arguments: " + Arrays.toString(args));

    DefaultOptionBuilder obuilder = new DefaultOptionBuilder();
    ArgumentBuilder abuilder = new ArgumentBuilder();
    GroupBuilder gbuilder = new GroupBuilder();

    Option inputOpt = obuilder.withLongName("input").withRequired(true)
            .withArgument(abuilder.withName("input").withMinimum(1).withMaximum(1).create())
            .withDescription("Input file").withShortName("i").create();

    Option modelOpt = obuilder.withLongName("model").withRequired(true)
            .withArgument(abuilder.withName("model").withMinimum(1).withMaximum(1).create())
            .withDescription("Model to use when classifying data").withShortName("m").create();

    Option helpOpt = obuilder.withLongName("help").withDescription("Print out help").withShortName("h")
            .create();//from  ww w.j a  v a2  s.co  m

    Group group = gbuilder.withName("Options").withOption(inputOpt).withOption(modelOpt).withOption(helpOpt)
            .create();

    try {
        Parser parser = new Parser();
        parser.setGroup(group);
        CommandLine cmdLine = parser.parse(args);

        if (cmdLine.hasOption(helpOpt)) {
            CommandLineUtil.printHelp(group);
            return;
        }

        File inputFile = new File(cmdLine.getValue(inputOpt).toString());

        if (!inputFile.isFile()) {
            throw new IllegalArgumentException(inputFile + " does not exist or is not a file");
        }

        File modelDir = new File(cmdLine.getValue(modelOpt).toString());

        if (!modelDir.isDirectory()) {
            throw new IllegalArgumentException(modelDir + " does not exist or is not a directory");
        }

        BayesParameters p = new BayesParameters();
        p.set("basePath", modelDir.getCanonicalPath());
        Datastore ds = new InMemoryBayesDatastore(p);
        Algorithm a = new BayesAlgorithm();
        ClassifierContext ctx = new ClassifierContext(a, ds);
        ctx.initialize();

        //TODO: make the analyzer configurable
        StandardAnalyzer analyzer = new StandardAnalyzer(Version.LUCENE_36);
        TokenStream ts = analyzer.tokenStream(null,
                new InputStreamReader(new FileInputStream(inputFile), "UTF-8"));

        ArrayList<String> tokens = new ArrayList<String>(1000);
        while (ts.incrementToken()) {
            tokens.add(ts.getAttribute(CharTermAttribute.class).toString());
        }
        String[] document = tokens.toArray(new String[tokens.size()]);

        ClassifierResult[] cr = ctx.classifyDocument(document, "unknown", 5);

        for (ClassifierResult r : cr) {
            System.err.println(r.getLabel() + "\t" + r.getScore());
        }
    } catch (OptionException e) {
        log.error("Exception", e);
        CommandLineUtil.printHelp(group);
    } catch (IOException e) {
        log.error("IOException", e);
    } catch (InvalidDatastoreException e) {
        log.error("InvalidDataStoreException", e);
    } finally {

    }
}

---