IkAnalyzer分词

package com.test;

import java.io.IOException;
import java.io.Reader;
import java.io.StringReader;
import java.util.Arrays;
import java.util.HashMap;
import java.util.LinkedHashMap;
import java.util.LinkedList;
import java.util.List;
import java.util.Map;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;
import org.wltea.analyzer.core.IKSegmenter;
import org.wltea.analyzer.core.Lexeme;

import com.test.entity.ArticleThesaurus;

public class IkAnalyzerTest {
    public static void main(String[] args) {
//      String str = " 铜管方<a href='http://auto.ifeng.com/' target='_blank'>< font color='#004276'>面</font></a>，3月铜管企业开工率为85.54%，4月达 88%。据了解，由于3月铜价低迷，铜管企业提前备货，精铜需求提前放量。"
//              + "3月铜杆线企业开工率为72.47%，4月上升至76%。开工率上行，"
//              + "但前期备货并不充足，导致现货市场集中采购增加，供应紧张凸显。fdsf紫铜铜inf的说法都是紫铜,"
//              + "我勒个去黄铜恩恩黄铜哈哈哈黄铜铜,我勒个去白铜，范德萨范德萨发白铜，古代有很多人用青铜器，是的就是这个东西它的名字是一种金属叫青铜。";
        //System.out.println("Jane Campion directed \"The Piano\" in 1993.");
        String str="由于对经济形势的的担忧,市场一直低迷,各方面消息被吸<a href='http://www.test.cn/quote/'>铜价</a> 收,市场经济困难局面早已被认同," +
                "随着各国政策的出台实施及各经济数据的公布,担忧情绪渐缓,沪铝即将迎来谷底反弹的时机。
  </p> <p> 操作思路<img src=\"####\" alt=\"锻铜铍铜\" />" +
                "15500<a href='http://www.test.cn/product/tjthj_ht/'>黄铜</a>锻铜以下紫铜板多单继续持有,15575以下果断购入多单,止损15250价位," +
                "如果突破15750价位并站稳,可继续加仓购入多单。<img src=\"www.baidu.com\" alt=\"范德萨发生的\" />";
        Pattern p = Pattern.compile("</?(A|a)(\n|.)*?>");
        Matcher m = p.matcher(str);
        str = m.replaceAll("");
        System.out.println("清除所有a标签:"+str);


        System.out.println("分词后："+ikAnalyzer(str));
        String afterFcStr = ikAnalyzer(str); // 分词后的字符串

        // 计算词频
        Map<String, Integer> words = new HashMap<String, Integer>();
        IKSegmenter seg = new IKSegmenter(new StringReader(afterFcStr), true);
        try {
            Lexeme l = null;
            while ((l = seg.next()) != null) {
                if (words.containsKey(l.getLexemeText()))
                    words.put(l.getLexemeText(),
                            words.get(l.getLexemeText()) + 1);
                else
                    words.put(l.getLexemeText(), 1);
            }
        } catch (IOException e) {
            e.printStackTrace();
        }

         for (Map.Entry<String, Integer> entry : words.entrySet()) {
         System.out.println("key= " + entry.getKey() + " and value= "
         + entry.getValue());
         }

         Integer count=words.get("铜价");
         if(count!=null){
         System.out.println("该词频："+count);
         }else{
         System.out.println("该词频不存在");
         }
    //添加文章内链 一篇文章不超过五个内链 多个关键词 只替换一个关键词
        List<ArticleThesaurus> listKeyWord = new LinkedList<ArticleThesaurus>();
        ArticleThesaurus at1 = new ArticleThesaurus("铜", "http://www.test.cn");
        ArticleThesaurus at2 = new ArticleThesaurus("铜价","http://www.test.cn/quote/");
        ArticleThesaurus at3 = new ArticleThesaurus("紫铜",
                "http://www.test.cn/product/tjthj_ct_zt/");
        ArticleThesaurus at4 = new ArticleThesaurus("黄铜",
                "http://www.test.cn/product/tjthj_ht/");
        ArticleThesaurus at5 = new ArticleThesaurus("白铜",
                "http://www.test.cn/product/tjthj_bt/");
        ArticleThesaurus at6 = new ArticleThesaurus("青铜",
                "http://www.test.cn/product/tjthj_qt/ ");
        listKeyWord.add(at1);
        listKeyWord.add(at2);
        listKeyWord.add(at3);
        listKeyWord.add(at4);
        listKeyWord.add(at5);
        listKeyWord.add(at6);

        String newStr ;
        newStr = afterFcStr;
        String article[] = afterFcStr.split("\\|");
        int successcount = 0;
        for (int i = 0; i < listKeyWord.size(); i++) {
            if (successcount == 5) {
                break;
            }
            String wordname = listKeyWord.get(i).getWord();
            Map<String, Integer> map = new LinkedHashMap<String, Integer>(); // 防止重复添加内链
            for (int j = 0; j < article.length; j++) {

                if (wordname.equals(article[j])) {
                    if (map.get(wordname)== null) {
                        map.put(wordname, 1);
                        Arrays.fill(article, j, j + 1, "<a href='"
                                + listKeyWord.get(i).getUrl() + "'>" + wordname
                                + "</a>");
                        successcount++;
                    }
                }

            }
        }

        // for(int i=0;i<listKeyWord.size();i++){
        // String wordname=listKeyWord.get(i).getWord();
        // Integer count=words.get(wordname);
        // if(successcount==5){
        // break;
        // }
        // if(count!=null){
        // //System.out.println("该词频："+count);
        // newStr=newStr.replaceFirst(wordname,
        // "<a href='"+listKeyWord.get(i).getUrl()+"'>"+wordname+"</a>");
        // successcount++;
        // }else{
        // //System.out.println("该词频不存在");
        // }
        // }
        System.out.println("内链优化后的文章:" + Arrays.toString(article));
        StringBuilder StrArticle=new StringBuilder();
        for(int i=0;i<article.length;i++){
            StrArticle.append(article[i]);
        }
        //System.out.println("被优化多少个内链:"+successcount);
        //System.out.println("内链优化后的文章字符串:" + StrArticle);
        String endStr=StrArticle.toString();
        if(successcount==0){    //可能分词导致部分关键词没有匹配到 则采用绝对字符匹配
            for (int i = 0; i < listKeyWord.size(); i++) {
                //判断文章里的超链接数
                int acount=occurTimes(endStr,"href=");
                if(acount==5){
                    break;
                }
                String wordname = listKeyWord.get(i).getWord();
                endStr=endStr.replaceFirst(wordname, "<a href='"
                                + listKeyWord.get(i).getUrl() + "'>" + wordname
                                + "</a>");

            }
        }

        //去除alt标签内的a内链
        System.out.println("内链优化后的文章字符串:"+endStr);
        Document doc = Jsoup.parseBodyFragment(endStr); // or Jsoup.parse(...);
        Elements images = doc.select("img");

        List<String> listAltStr=new LinkedList<String>();

        for(Element image : images){
           // System.out.printf("%s:%s%n", image.attr("src"), image.attr("alt"));
            //System.out.println(image.attr("alt"));

            String altStr=image.attr("alt");
            Pattern p1 = Pattern.compile("</?(A|a)(\n|.)*?>");
            Matcher m1 = p.matcher(altStr);
            altStr = m1.replaceAll("");
            listAltStr.add(altStr);
            image.attr("alt", altStr);

            //System.out.println(altStr);
        }
        doc.select("img").listIterator();
        System.out.println("end内链优化后的文章字符串:" + doc.select("body").html());
    }

    /**
     * 字符在字符串中出现的次数
     *
     * @param string
     * @param a
     * @return
     */
    public static int occurTimes(String string, String a) {
        int pos = -2;
        int n = 0;

        while (pos != -1) {
            if (pos == -2) {
                pos = -1;
            }
            pos = string.indexOf(a, pos + 1);
            if (pos != -1) {
                n++;
            }
        }
        return n;
    }

    public static String ikAnalyzer(String str) {

        Reader input = new StringReader(str);
        // 智能分词关闭（对分词的精度影响很大）
        IKSegmenter iks = new IKSegmenter(input, true);
        Lexeme lexeme = null;
        StringBuilder sb = new StringBuilder();

        try {
            while ((lexeme = iks.next()) != null) {

                sb.append(lexeme.getLexemeText()).append("|");
            }
        } catch (IOException e) {
            e.printStackTrace();
        }

        return sb.toString();
    }

}

 

原文链接地址：http://skyfar666.iteye.com/blog/2087029