dex.java

这是一个matlab的java实现。里面有许多内容。请大家慢慢捉摸。

/* 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. */

/** Driver file for DEX: automatically extracting social networks and
 *   contact information from email and the web.
 *   @author Aron Culotta <A HREF="mailto:[e-mail]">[e-mail]</A>
*/

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

import edu.umass.cs.mallet.projects.dex.types.*;
import edu.umass.cs.mallet.projects.dex.graph.*;
import edu.umass.cs.mallet.base.util.*;
import java.io.*;
import java.util.logging.*;


public class DEX {

	private static Logger logger = MalletLogger.getLogger(DEX.class.getName());

	public static void main (String[] args) {
		commandOptions.process (args);
		commandOptions.logOptions (logger);
		makeDir(outputDirOption.value);
		PeopleManager pm = new PeopleManager (inputDirOption.value, outputDirOption.value,
																					stopListOption.value, userNameOption.value,
																					readPeopleOption.value, vcfOutputOption.value,
																					htmlOutputOption.value);
		pm.writeTXT (new File ("people.txt"));
		pm.expandSocialNetwork (numberHopsOption.value, crfFileOption.value);
		if (writePeopleOption.value != null)
			pm.writePeople (writePeopleOption.value);
		pm.writeTXT (new File ("people.txt"));
	  pm.writeHTML (htmlOutputOption.value);
		pm.writeVCF (vcfOutputOption.value);
		PeopleGraph pg = new PeopleGraph (pm.getPeople());
		pg.printAllStatistics ();		
	}

	public static void makeDir(File dir) {
		try {
	    if(dir.exists() == false)
				dir.mkdir();
		} catch (SecurityException e) {
	    System.out.println("No permission to make directory " + dir);
		}
	}

	/** options */
	static CommandOption.File inputDirOption = new CommandOption.File
	(DEX.class, "input-dir", "FILE", true, null, "input directory containing emails", null);
	
	static CommandOption.File outputDirOption = new CommandOption.File
	(DEX.class, "output-dir", "FILE", true, null, "output directory to place discovered web pages", null);
	
	static CommandOption.File csvOutputOption = new CommandOption.File
	(DEX.class, "csv-output", "FILE", true, new File ("output.csv"), "csv output file", null);
	
 	static CommandOption.File htmlOutputOption = new CommandOption.File
	(DEX.class, "html-output", "FILE", true, new File ("output.html"), "html output file", null);

 	static CommandOption.File vcfOutputOption = new CommandOption.File
	(DEX.class, "vcf-output", "FILE", true, new File ("output.vcf"), "VCF (vCard format) output file", null);
    
	static CommandOption.File stopListOption = new CommandOption.File
	(DEX.class, "stop-list", "FILE", true, new File ("/usr/col/tmp1/ronb/mallet/exp/calo/stop"), "stop list file for keywords", null);
	
	static CommandOption.File readPeopleOption = new CommandOption.File
	(DEX.class, "read-people", "FILE", true, null, "file to read People", null);
	
	static CommandOption.File writePeopleOption = new CommandOption.File
	(DEX.class, "write-people", "FILE", true, null, "file to write People", null);
	
	static CommandOption.String userNameOption = new CommandOption.String
	(DEX.class, "user-name", "FILE", true, null, "name of user", null);
	
	static CommandOption.File crfFileOption = new CommandOption.File
	(DEX.class, "crf-file", "FILE", true, new File ("/usr/col/tmp1/culotta/mallet/exp/address/crf.obj"), "CRF object file to perform contact record extraction", null);
	
	static CommandOption.Boolean confidencePredictionOption = new CommandOption.Boolean
	(DEX.class, "confidence-prediction", "true|false", false, false,
	 "predict confidence of each extracted field?", null);
	
	static CommandOption.Boolean searchWebOption = new CommandOption.Boolean
	(DEX.class, "search-web", "true|false", true, true,
	 "True if we should query google. If false, assumes <outputDir> contains files for each person.", null);
	
	static CommandOption.Integer numberHopsOption = new CommandOption.Integer
	(DEX.class, "number-hops", "INTEGER", true, 2,
	 "The number of outlink hops to explore from the initial set of people.", null);
	
	static final CommandOption.List commandOptions =
	new CommandOption.List (
		"DEX: Automated Rolodex Construction from Email.",
		new CommandOption[] {
				    inputDirOption,
				    outputDirOption,
				    userNameOption,
				    confidencePredictionOption,
				    crfFileOption,
				    searchWebOption,
				    csvOutputOption,
				    htmlOutputOption,
						vcfOutputOption,
				    stopListOption,
				    numberHopsOption,
				    readPeopleOption,
				    writePeopleOption,
		});
}