利用simhash计算文本相似度
2024-10-15 12:25:13
摘自:http://www.programcreek.com/java-api-examples/index.php?source_dir=textmining-master/src/com/gta/simhash/SimHash.java
package com.gta.simhash; public class Test { public static void main(String[] args) {
// TODO Auto-generated method stub String s3 = "�������Ϻ���������죬���������������������Ͼ������ݣ����������ţ����ϣ��ൺ���人�����ݣ����ڣ��ɶ���������̫ԭ����ɳ�����֣�������֣�ݣ���������������³ľ�룬���ݣ��������Ϸʣ��ߺ�";
String s4 = "�������Ϻ���������죬���������������������Ͼ������ݣ����������ţ����ϣ��ൺ���人�����ݣ����ڣ��ɶ���������̫ԭ����ɳ�����֣�������֣�ݣ�����";
SimHash hash1 = new SimHash(s3, 64, 8);
SimHash hash2 = new SimHash(s4, 64, 8);
hash1.getResult(hash2);
} }
package com.gta.simhash; import java.io.IOException;
import java.math.BigInteger;
import java.util.List;
import java.util.ArrayList; import org.wltea.analyzer.lucene.IKAnalyzer;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.tokenattributes.CharTermAttribute; public class SimHash {
private String tokens;
private int hashBits = 64;
private int distance = 5; public SimHash(String tokens)
{
this.tokens = tokens;
} public SimHash(String tokens, int hashBits, int distance)
{
this.tokens = tokens;
this.hashBits = hashBits;
this.distance = distance;
} public List<TermDict> tokenizer()
{
List<TermDict> terms = new ArrayList<TermDict>();
IKAnalyzer analyzer = new IKAnalyzer(true);
try {
TokenStream stream = analyzer.tokenStream("", this.tokens);
CharTermAttribute cta = stream.addAttribute(CharTermAttribute.class);
stream.reset();
int index = -1;
while (stream.incrementToken())
{
if ((index = isContain(cta.toString(), terms)) >= 0)
{
terms.get(index).setFreq(terms.get(index).getFreq()+1);
}
else
{
terms.add(new TermDict(cta.toString(), 1));
}
}
analyzer.close();
} catch (IOException e) {
e.printStackTrace();
}
return terms;
} public int isContain(String str, List<TermDict> terms)
{
for (TermDict td : terms)
{
if (str.equals(td.getTerm()))
{
return terms.indexOf(td);
}
}
return -1;
} public BigInteger simHash(List<TermDict> terms)
{
int []v = new int[hashBits];
for (TermDict td : terms)
{
String str = td.getTerm();
int weight = td.getFreq();
BigInteger bt = shiftHash(str);
for (int i = 0; i < hashBits; i++)
{
BigInteger bitmask = new BigInteger("1").shiftLeft(i);
if ( bt.and(bitmask).signum() != 0)
{
v[i] += weight;
}
else
{
v[i] -= weight;
}
}
} BigInteger fingerPrint = new BigInteger("0");
for (int i = 0; i < hashBits; i++)
{
if (v[i] >= 0)
{
fingerPrint = fingerPrint.add(new BigInteger("1").shiftLeft(i)); // update the correct fingerPrint
}
}
return fingerPrint;
} public BigInteger shiftHash(String str)
{
if (str == null || str.length() == 0)
{
return new BigInteger("0");
}
else
{
char[] sourceArray = str.toCharArray();
BigInteger x = BigInteger.valueOf((long) sourceArray[0] << 7);
BigInteger m = new BigInteger("131313");
for (char item : sourceArray)
{
x = x.multiply(m).add(BigInteger.valueOf((long)item));
}
BigInteger mask = new BigInteger("2").pow(hashBits).subtract(new BigInteger("1"));
boolean flag = true;
for (char item : sourceArray)
{
if (flag)
{
BigInteger tmp = BigInteger.valueOf((long)item << 3);
x = x.multiply(m).xor(tmp).and(mask);
}
else
{
BigInteger tmp = BigInteger.valueOf((long)item >> 3);
x = x.multiply(m).xor(tmp).and(mask);
}
flag = !flag;
} if (x.equals(new BigInteger("-1")))
{
x = new BigInteger("-2");
}
return x;
}
} public BigInteger getSimHash()
{
return simHash(tokenizer());
} public int getHammingDistance(SimHash hashData)
{
BigInteger m = new BigInteger("1").shiftLeft(hashBits).subtract(new BigInteger("1"));
System.out.println(getFingerPrint(getSimHash().toString(2)));
System.out.println(getFingerPrint(hashData.getSimHash().toString(2)));
BigInteger x = getSimHash().xor(hashData.getSimHash()).and(m);
int tot = 0;
while (x.signum() != 0)
{
tot += 1;
x = x.and(x.subtract(new BigInteger("1")));
}
System.out.println(tot);
return tot;
} public String getFingerPrint(String str)
{
int len = str.length();
for (int i = 0; i < hashBits; i++)
{
if (i >= len)
{
str = "0" + str;
}
}
return str;
} public void getResult(SimHash hashData)
{
if (getHammingDistance(hashData) <= distance)
{
System.out.println("match");
}
else
{
System.out.println("false");
}
} }
最新文章
- 289. Game of Life -- In-place计算游戏的下一个状态
- PNG格式小图标的CSS任意颜色赋色技术
- linux下利用curl监控网页shell脚本
- 使用java实现持续移动的小球
- Sales_item
- WCf的理解
- io系统
- 页面动态数据的滚动效果——jquery滚动组件(vticker.js)
- 2机器学习实践笔记(k-最近邻)
- 小程序新能力-个人开发者尝鲜微信小程序
- 【BZOJ1012】【JSOI2008】最大数maxnumber
- SpringCloud---网关概念、Zuul项目搭建(六)
- Spark集群部署(standLone)模式
- 关于MySQL中的8个 character_set 变量
- node-sass 安装
- 【HDU 5647】DZY Loves Connecting(树DP)
- LINUX ORACLE 启动与关闭
- 黑盒测试用例设计——PICT
- 【朴灵评注】JavaScript 运行机制详解:再谈Event Loop
- 通讯协议(三)Protocol Buffers协议