de.tudarmstadt.ukp.wikipedia.util.GraphUtilities.java Source code

Introduction

Here is the source code for de.tudarmstadt.ukp.wikipedia.util.GraphUtilities.java

Source

/*******************************************************************************
 * Copyright (c) 2010 Torsten Zesch.
 * All rights reserved. This program and the accompanying materials
 * are made available under the terms of the GNU Lesser Public License v3
 * which accompanies this distribution, and is available at
 * http://www.gnu.org/licenses/lgpl.html
 * 
 * Contributors:
 *     Torsten Zesch - initial API and implementation
 ******************************************************************************/
package de.tudarmstadt.ukp.wikipedia.util;

import java.util.*;

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import de.tudarmstadt.ukp.wikipedia.api.Page;

public class GraphUtilities {

    private static final Log logger = LogFactory.getLog(GraphUtilities.class);

    public static Set<Integer> getRandomPageSubset(Iterable<Page> pages, int pResultSetSize) {
        Set<Integer> pageIDs = new HashSet<Integer>();
        while (pages.iterator().hasNext()) {
            pageIDs.add(pages.iterator().next().getPageId());
        }
        return getRandomPageSubset(pageIDs, pResultSetSize);
    }

    /** Get a random subset (of size pSize) of the page set passed to the method.
     * @param pPageIDs The pages.
     * @param pResultSetSize The size of the result set.
     * @return A random subset of the original page set of the given size or null, if the requested subset size is larger than the original page set.
     */
    public static Set<Integer> getRandomPageSubset(Set<Integer> pPageIDs, int pResultSetSize) {

        Set<Integer> uniqueRandomSet = new HashSet<Integer>();

        if (pPageIDs.size() < pResultSetSize) {
            logger.error("Requested subset size is larger than the original page set size.");
            return null;
        }

        Random rand = new Random();

        Object[] pageIdArray = pPageIDs.toArray();

        // If pSize is quite close to the size of the original pageSet the probability of generating the offset of the last missing pageIDs is quite low, with the consequence of unpredictable run-time.
        // => if more than the half of pages should be included in the result set, better remove random numbers than adding them
        if (pResultSetSize > (pPageIDs.size() / 2)) {
            uniqueRandomSet.addAll(pPageIDs);
            while (uniqueRandomSet.size() > pResultSetSize) {
                int randomOffset = rand.nextInt(pPageIDs.size());
                if (uniqueRandomSet.contains(pageIdArray[randomOffset])) {
                    uniqueRandomSet.remove(pageIdArray[randomOffset]);
                }
            }
        } else {
            while (uniqueRandomSet.size() < pResultSetSize) {
                int randomOffset = rand.nextInt(pPageIDs.size());
                if (!uniqueRandomSet.contains(pageIdArray[randomOffset])) {
                    uniqueRandomSet.add((Integer) pageIdArray[randomOffset]);
                }
            }
        }

        return uniqueRandomSet;
    }
}