欢迎您访问程序员文章站本站旨在为大家提供分享程序员计算机编程知识!
您现在的位置是: 首页

IkAnalyzer分词、词频、内链优化

程序员文章站 2022-05-17 18:54:12
...
IK 中文分词  版本 5.0
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();
	}

}



下面提供源码包下载 所用ik没有用到jar包 用到了源码包 方便调试和修改源代码

iteye传不上来
只能让大家去csdn下载了http://download.csdn.net/detail/koproblem/7575521
相关标签: ik