query.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. */

/**  Query to Google.
 *   @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 java.util.Vector;
import java.util.regex.*;

public class Query {
	public Query(int num) {
		minOccurrences = num;
	}
	
	public Vector buildQueriesWithDomainFields(Vector queries, String name,
																						 Vector domains, int numOfFields){
		Vector currentDomains = new Vector();
		for (int i = 0; i < domains.size(); i++) {
	    String domain = ( (CountedString) domains.elementAt(i)).str;
	    // Take three last fields of the domain
	    Pattern pat = Pattern.compile("\\.");
	    String[] fields = pat.split(domain);
	    domain = "";
	    if (fields.length >= numOfFields) {
				for(int j = numOfFields; j > 1; j--)
					domain = domain + fields[fields.length - j] + ".";
				domain = domain + fields[fields.length - 1];
				for (int ii = 0; ii < currentDomains.size(); ii++) {
					String currentDomain = (String)currentDomains.get(ii);
					if(currentDomain.equals(domain))
						continue;
				}
				currentDomains.add(domain);
				String query = "\"" + name + "\" site:" + domain;
				System.out.println("Query: " + query);
				queries.add(query);
	    }
		}
		return queries;
	}   
	
	public Vector buildQueries(Person person) {
		Vector queries = new Vector();
		if(person.getNumberOfOccurrences() <= minOccurrences) {
	    //We don't work with rarely occurred people
	    return queries;
		}
		if(person.names.size() == 0) {
	    //No name
	    return queries;
		}
		// Take only first 3 names in the list
		for(int i = 0; i < person.names.size() && i < 3; i++) {
	    String name = ( (CountedString) person.names.elementAt(i)).str;
	    queries = buildQueriesWithDomainFields(queries, name, person.domains, 3);
	    queries = buildQueriesWithDomainFields(queries, name, person.domains, 2);
		}
		// Now queries without domain
		for(int i = 0; i < person.names.size() && i < 3; i++) {
	    String name = ( (CountedString) person.names.elementAt(i)).str;
	    if(name.indexOf(" ") == -1)
		continue;
	    String query = "\"" + name + "\"";
	    System.out.println("Query: " + query);
	    queries.add(query);
		}
		// Only surname with domains
		String surname = person.getSurname();
		if(surname != null) {
	    queries = buildQueriesWithDomainFields(queries, surname, person.domains, 3);
	    //queries = buildQueriesWithDomainFields(queries, surname, person.domains, 2);
		}
		return queries;
	}
	
	//Fields
    private int minOccurrences;
}