| 发表于：2007-03-08 11:11:048楼 得分:0 |
我用过c#阿 但是这个比较急我没法再去学习java了
下面是其中一个类的程序
import java.io.bufferedreader;
import java.io.filenotfoundexception;
import java.io.filereader;
import java.io.ioexception;
import java.util.iterator;
import java.util.linkedlist;
import java.util.listiterator;
import java.util.hashmap;
import java.util.set;
import java.util.stringtokenizer;
import java.util.arraylist;
//author=xingrui ji
//data=oct 7 2006
public class support {
// reads sentence
public static linkedlist readsentences（string filename)
throws filenotfoundexception, ioexception {
filereader myfilereader = new filereader（filename);
bufferedreader mybufferedreader = new bufferedreader（myfilereader);
string line = mybufferedreader.readline（);
linkedlist sentencelist = new linkedlist（);
while （line != null) {
sentencelist.add（line);
line = mybufferedreader.readline（);
}
return sentencelist;
}
// get every word 's tag
static hashmap addwordtag（hashmap wordmap, string keyword) {
keyword = keyword.tolowercase（);
string[] tagword = keyword.split（ "/ ");
if （tagword.length < 2)
return wordmap;
if （wordmap.containskey（tagword[1])) {
arraylist arrattr = （arraylist) wordmap.get（tagword[1]);
if （!arrattr.contains（tagword[0])) {
arrattr.add（tagword[0]);
}
} else {
arraylist arrattr = new arraylist（);
arrattr.add（tagword[0]);
wordmap.put（tagword[1], arrattr);
}
return wordmap;
}
// support function, add a new key to hashmap or increase value as count
static hashmap addwordinc（hashmap wordmap, string keyword)
{
keyword = keyword.tolowercase（);
if （wordmap.containskey（keyword)) {
integer wordcounterinteger = （integer) wordmap.get（keyword);
int wordcounterint = wordcounterinteger.intvalue（);
wordcounterinteger = new integer（wordcounterint + 1);
wordmap.put（keyword, wordcounterinteger);
} else {
wordmap.put（keyword, new integer（1));
}
return wordmap;
}
// get every word 's tag
static hashmap getwordtag（linkedlist sentencelist) {
listiterator i = sentencelist.listiterator（);
hashmap wordmap = new hashmap（);
while （i.hasnext（)) {
string currentsentence = （string) i.next（);
stringtokenizer sentencetokenizer = new stringtokenizer（
currentsentence);
while （sentencetokenizer.hasmoretokens（)) {
string keyword = sentencetokenizer.nexttoken（).tolowercase（);
wordmap = addwordtag（wordmap, keyword);
}
}
return wordmap;
}
// get every tag 's count
static hashmap counttagsingle（linkedlist sentencelist)
{
listiterator i = sentencelist.listiterator（);
hashmap wordmap = new hashmap（);
while （i.hasnext（)) {
string currentsentence = （string) i.next（);
stringtokenizer sentencetokenizer = new stringtokenizer（
currentsentence);
while （sentencetokenizer.hasmoretokens（)) {
string keyword = sentencetokenizer.nexttoken（).tolowercase（);
string[] attrword = keyword.split（ "/ ");
if （attrword.length == 1)
break;
wordmap = addwordinc（wordmap, attrword[0]);
}
}
return wordmap;
}
// counts p（x,y)
static hashmap counttagpair（linkedlist sentencelist)
{
listiterator i = sentencelist.listiterator（);
hashmap wordmap = new hashmap（);
while （i.hasnext（)) {
string currentsentence = （string) i.next（);
stringtokenizer sentencetokenizer = new stringtokenizer（
currentsentence);
if （sentencetokenizer.hasmoretokens（)) {
string word1 = sentencetokenizer.nexttoken（).tolowercase（);
string[] attrword1 = word1.split（ "/ ");
if （attrword1.length == 1)
break;
while （sentencetokenizer.hasmoretokens（)) {
string word2 = sentencetokenizer.nexttoken（).tolowercase（);
attrword1 = word1.split（ "/ ");
string[] attrword2 = word2.split（ "/ ");
if （attrword2.length == 1)
break;
string keyword = attrword1[0] + " " + attrword2[0];
wordmap = addwordinc（wordmap, keyword);
word1 = word2;
}
}
}
return wordmap;
}
// calculates p（w_x ¦ w_y) and stores in hashmap as value
// p（wx ¦wy)=p（wx,wy)/p（wy)
public static hashmap calcposterior（hashmap tagtimesmap,
hashmap tagpairtimesmap) {
hashmap probmap = new hashmap（);
set wordset = tagpairtimesmap.keyset（);
iterator itword = wordset.iterator（);
while （itword.hasnext（)) {
string keyword = （string) itword.next（);
stringtokenizer sentencetokenizer = new stringtokenizer（keyword);
if （sentencetokenizer.hasmoretokens（)) {
string word1 = sentencetokenizer.nexttoken（).tolowercase（);
integer freqword = （integer) tagpairtimesmap.get（keyword);
double probword = freqword.doublevalue（)
/ （（integer) tagtimesmap.get（word1)).doublevalue（);
probmap.put（keyword, new float（probword));
}
}
return probmap;
}
}
| | |
|
