hanlp分词

首先在pom文件中加入分词处理的JAR包

		<dependency>
			<groupId>com.hankcs</groupId>
			<artifactId>hanlp</artifactId>
			<version>portable-1.6.1</version>
		</dependency>

1、创建封装分词数据的Bean

package com.qlys.frame.model.impl;

import java.io.Serializable;
import java.util.concurrent.atomic.AtomicInteger;

/**
 * 分词基本信息
 */
public class SegmentWord implements Serializable {

	/**
	 * 
	 */
	private static final long serialVersionUID = 5662341029767237202L;

	// 词名
    private String name;
    
    // 词性
    private String pos;
    
    // 词频
    private AtomicInteger frequency = new AtomicInteger();
    
    public SegmentWord(String name,String pos) {
    	this.name = name;
    	this.pos = pos;
    }

	public String getName() {
		return name;
	}

	public void setName(String name) {
		this.name = name;
	}

	public String getPos() {
		return pos;
	}

	public void setPos(String pos) {
		this.pos = pos;
	}

	public AtomicInteger getFrequency() {
		return frequency;
	}

	public void setFrequency(AtomicInteger frequency) {
		this.frequency = frequency;
	}
    
    
    
    
}

2、进行分词并形成词频信息

package com.qlys.frame.util;


import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.Optional;
import java.util.concurrent.ConcurrentHashMap;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import com.hankcs.hanlp.corpus.tag.Nature;
import com.hankcs.hanlp.seg.common.Term;
import com.hankcs.hanlp.tokenizer.NLPTokenizer;
import com.qlys.frame.model.impl.SegmentWord;
/**
 * 分词处理工具类
 */
public class TokenizerUtil {

	
	
	private static final Logger log = LoggerFactory.getLogger(TokenizerUtil.class);

	/**
	 * 分词策略 
	 * @return 分词信息
	 */
	public List<Term> segmentCategory(String content){
		return NLPTokenizer.segment(content);
	}
	
	/**
	 * 分词并计算词频
	 * @param content 需要进行分词的文本
	 * @return 分词后的词组信息及词频
	 */
	public Map<String, SegmentWord> segment(String content) {
		log.debug("开始执行分词");
		List<Term> termList = this.segmentCategory(content);
		Map<String, SegmentWord> map = new ConcurrentHashMap<String, SegmentWord>();
		termList.forEach(i -> Optional.ofNullable(i.nature == Nature.w ? null : i.nature).ifPresent(m -> map
				.computeIfAbsent(i.word, k -> new SegmentWord(i.word, i.nature.toString())).getFrequency().incrementAndGet()));
		return map;
	}
	/**
	 * 相似度运算
	 *
	 * @param s 分词1
	 * @param o 分词2
	 * @return 分词1和分词2的相似度
	 */
	public double similarity(Map<String, SegmentWord> s,Map<String, SegmentWord> o) {
		List<String> keys = new ArrayList<String>();
		keys.addAll(s.keySet());
		keys.retainAll(o.keySet());
		//运算分子数据
		return keys.stream().map(val->s.get(val).getFrequency().intValue()*o.get(val).getFrequency().intValue()).reduce((a,b)->a+b).get()
		/
		Math.sqrt(s.values().stream().map(val->Math.pow(val.getFrequency().intValue(),2)).reduce((a,b)->a+b).get())
		/ Math.sqrt(o.values().stream().map(val->Math.pow(val.getFrequency().intValue(),2)).reduce((a,b)->a+b).get());
	}

	
	public static void main(String[] args) {
		TokenizerUtil util = new TokenizerUtil();
		Map<String, SegmentWord> map = util.segment("我们把香蕉给猴子因为它们饿了");
		Map<String, SegmentWord> map1 = util.segment("我们不能把香蕉给猴子因为它们还没有成熟");
		
		System.out.println(util.similarity(map, map1));
		System.out.println(util.similarity(util.segment("abc123"), util.segment("abc")));
	}
	
}