homepagefinder.java

常用机器学习算法,java编写源代码,内含常用分类算法,包括说明文档

/* Copyright (C) 2003 Univ. of Massachusetts Amherst, Computer Science Dept.
   This file is part of "MALLET" (MAchine Learning for LanguagE Toolkit).
   http://www.cs.umass.edu/~mccallum/mallet
   This software is provided under the terms of the Common Public License,
   version 1.0, as published by http://www.opensource.org.  For further
   information, see the file `LICENSE' included with this distribution. */

/**  Generate queries and search for people's homepages on the web.
 *   @author Ron Bekkerman <A HREF="mailto:[e-mail]">[e-mail]</A>
*/

package edu.umass.cs.mallet.projects.dex.web;

import edu.umass.cs.mallet.projects.dex.types.*;
import edu.umass.cs.mallet.base.util.MalletLogger;

import java.util.logging.*;
import java.util.*;
import java.io.*;

public class HomePageFinder {
	private static Logger logger = MalletLogger.getLogger(HomePageFinder.class.getName());
	public final static int  MIN_NUM_OCCURRENCES = 0;
	File outputDir;
	String userName;
	HashSet stopWords;
	
	public HomePageFinder (File outputDir, String userName, HashSet stopWords) {
		this.outputDir = outputDir;
		this.userName = userName;
		this.stopWords = stopWords;		
	}

	public void findHomePagesFor (People people) {
		Vector threads = new Vector();
		Iterator piter = people.iterator();
		while (piter.hasNext()) {
	    Person person = (Person)piter.next();
	    if(person.processedForWebPages) // Processed on previous stages
				continue;
	    if(person.findLogin(userName)) // No need in looking for email directory user
				continue;
	    Query q = new Query(MIN_NUM_OCCURRENCES); // Look for people who appear more than 0 times
	    Vector queries = q.buildQueries(person);
	    if(queries.size() == 0){
				logger.info("No queries generated for ");
				person.printPersonalInfo();
				continue;
	    }
	    String destDirName = generateOutDirName(outputDir.toString(), person);
	    makeDir(destDirName);
	    WebThread thread = new WebThread(queries, destDirName, person, stopWords);
	    threads.addElement(thread);
	    logger.fine(thread.getName() + " added; Threads size: " + threads.size());
	    logger.fine("Active count: " + Thread.activeCount());
	    person.processedForWebPages = true;
	    if(Thread.activeCount() > 63) {
				logger.info("Too many threads in the system - wait");
				waitForAllThreadsDead(threads);
	    }
	    logger.fine("Thread " + thread.getName() + " is about to start");
	    thread.start();
		}
		waitForAllThreadsDead(threads);
	}

	private void waitForAllThreadsDead(Vector threads) {
		for (int i = 0; i < threads.size(); i++) {
	    try{
				WebThread thread = (WebThread) threads.elementAt(i);
				logger.fine (thread.getName() + ") Wait for thread to finish");
				thread.join();
				logger.fine (thread.getName() + ") Thread finished");
	    } catch (InterruptedException err){
				logger.warning("Problem with interruptions");
	    }
		}
	}
	
	private void makeDir(String dir) {
		File outDir = new File(dir);
		try {
	    if(outDir.exists() == false)
				outDir.mkdir();
		} catch (SecurityException e) {
	    logger.warning("No permission to make directory " + dir);
		}
	}

	public static String generateOutDirName(String outDir, Person person) {
		return outDir + File.separator +
	    ((CountedString)person.names.elementAt(0)).str;
	}
	
}