org.dkpro.similarity.experiments.sts2013.util.Features2Arff.java Source code

Introduction

Here is the source code for org.dkpro.similarity.experiments.sts2013.util.Features2Arff.java

Source

/*******************************************************************************
 * Copyright 2013
 * Ubiquitous Knowledge Processing (UKP) Lab
 * Technische Universitt Darmstadt
 *
 * All rights reserved. This program and the accompanying materials
 * are made available under the terms of the GNU Public License v3.0
 * which accompanies this distribution, and is available at
 * http://www.gnu.org/licenses/gpl-3.0.txt
 ******************************************************************************/
package org.dkpro.similarity.experiments.sts2013.util;

import java.io.File;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Collection;
import java.util.HashMap;
import java.util.List;
import java.util.Map;

import org.apache.commons.io.FileUtils;
import org.apache.commons.lang.StringUtils;
import org.dkpro.similarity.experiments.sts2013.Pipeline.Dataset;
import org.dkpro.similarity.experiments.sts2013.Pipeline.Mode;
import org.springframework.core.io.Resource;
import org.springframework.core.io.support.PathMatchingResourcePatternResolver;

import static org.dkpro.similarity.experiments.sts2013.Pipeline.FEATURES_DIR;
import static org.dkpro.similarity.experiments.sts2013.Pipeline.GOLDSTANDARD_DIR;
import static org.dkpro.similarity.experiments.sts2013.Pipeline.MODELS_DIR;

public class Features2Arff {
    public static final String LF = System.getProperty("line.separator");

    public static void toArffFile(Mode mode, Dataset... datasets) throws IOException {
        for (Dataset dataset : datasets) {
            String path = GOLDSTANDARD_DIR + "/" + mode.toString().toLowerCase() + "/STS.gs." + dataset.toString()
                    + ".txt";

            PathMatchingResourcePatternResolver r = new PathMatchingResourcePatternResolver();
            Resource res = r.getResource(path);

            toArffFile(mode, dataset, res.getFile());
        }
    }

    @SuppressWarnings("unchecked")
    private static void toArffFile(Mode mode, Dataset dataset, File goldStandard) throws IOException {
        System.out.println("Generating ARFF file");

        Collection<File> files = FileUtils.listFiles(
                new File(FEATURES_DIR + "/" + mode.toString().toLowerCase() + "/" + dataset.toString()),
                new String[] { "txt" }, true);

        String arffString = toArffString(files, goldStandard);

        FileUtils.writeStringToFile(
                new File(MODELS_DIR + "/" + mode.toString().toLowerCase() + "/" + dataset.toString() + ".arff"),
                arffString);

        System.out.println(" - done");
    }

    @SuppressWarnings("unchecked")
    private static String toArffString(Collection<File> csvFiles, File goldFile) throws IOException {
        // Create the Arff header
        StringBuilder arff = new StringBuilder();
        arff.append("@relation temp-relation" + LF);
        arff.append(LF);

        // Init data object
        Map<Integer, List<Double>> data = new HashMap<Integer, List<Double>>();

        for (File file : csvFiles) {
            String feature = file.getParentFile().getName() + "/"
                    + file.getName().substring(0, file.getName().length() - 4);
            // feature = feature.replaceAll(",", "");

            // Add the attribute to the Arff header
            arff.append("@attribute " + feature + " numeric" + LF);

            // Read data
            List<String> lines = FileUtils.readLines(file);
            for (int doc = 1; doc <= lines.size(); doc++) {
                String line = lines.get(doc - 1);

                if (line.length() > 0) // Ignore empty lines
                {
                    double value = Double.parseDouble(line); // There's just the score on the line, nothing else.

                    // Limit to [0;5] interval
                    if (value > 5.0)
                        value = 5.0;
                    if (value < 0.0)
                        value = 0.0;

                    // Get doc object in data list
                    List<Double> docObj;
                    if (data.containsKey(doc))
                        docObj = data.get(doc);
                    else
                        docObj = new ArrayList<Double>();

                    // Put data
                    docObj.add(value);
                    data.put(doc, docObj);
                }
            }
        }

        // Add gold attribute to attribute list in header
        // We also need to do this for unlabeled data
        arff.append("@attribute gold real" + LF);

        // Add gold similarity score 
        List<String> lines;
        if (goldFile != null) {
            lines = FileUtils.readLines(goldFile);
        } else {
            lines = new ArrayList<String>();
            for (int i = 0; i < FileUtils.readLines(csvFiles.iterator().next()).size(); i++)
                lines.add("0.0");
        }

        for (int doc = 1; doc <= lines.size(); doc++) {
            System.out.println(lines.get(doc - 1).length());
            System.out.println(lines.get(doc - 1));
            if (lines.get(doc - 1).length() == 0) {
                System.out.println("here2");
                break;
            }

            double value = Double.parseDouble(lines.get(doc - 1));

            System.out.println(doc);

            List<Double> docObj = data.get(doc);
            docObj.add(value);
            data.put(doc, docObj);
        }

        // Finalize header
        arff.append(LF);
        arff.append("@data" + LF);

        // Write data
        for (int i = 1; i <= data.keySet().size(); i++) {
            String dataItem = StringUtils.join(data.get(i), ",");

            arff.append(dataItem + LF);
        }

        return arff.toString();
    }
}