java项目Kanalyzer分词器与mysql配合使用 分词开源库

快十年没碰过编程了，最近公司服务器换了一下，挪腾机器的过程里面顺便触发了再写点程序玩玩的想法，于是拿起.net的教材看了两天，练手写了个，测试反正能用，哈哈，发来让高手指点指点，揪揪错！

盘古开源分词组件直接去http:///下载就是了，Pangu.dll和Pangu.xml文件放入wwwroot的bin目录，字典别忘记放进去bin下的Dictionaries目录，呵呵，pangu.xml要设置好Dictionaries目录位置

 

用aspx共写了三个程序，第一个：

文件名：Default.aspx，存放目录：wwwroot

 

代码

<%        @ Page Language 
       = 
       " 
       C# 
       " 
        AutoEventWireup 
       = 
       " 
       true 
       " 
         CodeFile 
       = 
       " 
       Default.aspx.cs 
       " 
        Inherits 
       = 
       " 
       _Default 
       " 
         
       %> 
       
        < 
       html 
       > 
       
        < 
       head runat 
       = 
       " 
       server 
       " 
       > 
       
        < 
       title 
       > 
       WWW.RCSKY.NET 
       </ 
       title 
       > 
       
        </ 
       head 
       > 
       
        < 
       body 
       > 
       
            < 
       p 
       > 
       
            < 
       form id 
       = 
       " 
       form1 
       " 
        runat 
       = 
       " 
       server 
       " 
       > 
       
            < 
       div  
       class 
         
       = 
         
       " 
       align-center 
       " 
       > 
       
                < 
       p 
       > 
       原文： 
       < 
       asp:Label ID 
       = 
       " 
       fc_content 
       " 
        runat 
       = 
       " 
       server 
       " 
        Text 
       = 
       " 
       分词内容 
       " 
       ></ 
       asp:Label 
       > 
       
                < 
       p 
       > 
       分词结果： 
       < 
       br 
       >< 
       asp:Label ID 
       = 
       " 
       fc_result 
       " 
        runat 
       = 
       " 
       server 
       " 
        Text 
       = 
       " 
       分词结果 
       " 
       ></ 
       asp:Label 
       > 
       
                < 
       p 
       >< 
       asp:DataGrid id 
       = 
       " 
       Orign 
       " 
        runat 
       = 
       " 
       server 
       " 
        HeaderStyle 
       - 
       BackColor 
       = 
       " 
       #aaaadd 
       " 
        AlternatingItemStyle 
       - 
       BackColor 
       = 
       " 
       #eeeeee 
       " 
         
       /> 
       
                < 
       br  
       /> 
       
            </ 
       div 
       > 
       
            </ 
       form 
       > 
       
        </ 
       body 
       > 
       
        </ 
       html 
       >

第二个：default.aspx.cs，存放目录：wwwroot

 

代码

using System;
using System.Data;
using System.Data.OleDb;
using Rcsky.GetKeyword;public partial class _Default : System.Web.UI.Page 
{
    private void Page_Load(Object src,EventArgs e)
    {
   string MyConnString = "Provider=Microsoft.Jet.OLEDB.4.0; Data Source="+Server.MapPath("DatabaseDir/data.mdb");
   string strSel = "select * from db_table where id="+Request.QueryString["id"];
   DataSet ds = new DataSet();
   OleDbConnection MyConn = new OleDbConnection(MyConnString);
   OleDbDataAdapter MyAdapter = new OleDbDataAdapter(strSel,MyConn);
   OleDbCommandBuilder custCB = new OleDbCommandBuilder(MyAdapter);
   MyAdapter.Fill(ds,"TB_content");
   Orign.DataSource = ds;
   Orign.DataMember = "TB_content";
   Orign.DataBind();
   if (ds.Tables[0].Rows.Count > 0) {        
    DataRow dr=ds.Tables[0].Rows[0];
    fc_content.Text=dr["description"].ToString();//对db_table的description字段进行分词
    fc_result.Text = Segment.DoSegment(fc_content.Text);
   }
   GC.Collect();
    }
}

 

第三个程序：keyword.cs，存放目录：wwwroot/App_Code

 

 

代码

using         System;
        using         System.Collections;
        using         System.Collections.Generic;

        namespace         Rcsky.GetKeyword  {    
            public                 class 
        Segment   {
                public                 static 
         
       string 
        DoSegment( 
       string 
        keyWord)          
        {             
                    return         DoSegment(keyWord,         " 
       <br> 
       " 
       ); //分词输出的间隔符
        }         
                public                 static 
         
       string 
        DoSegment( 
       string 
        keyWord, 
       string 
        separator)         {               
            PanGu.Segment.Init();               
            PanGu.Segment segment         =                 new 
        PanGu.Segment();               
            ICollection        <        PanGu.WordInfo        > 
        words  
       = 
        segment.DoSegment(keyWord);               
            keyWord         =                 "" 
       ;               
                    int         i        = 
       0 
       ;             
                    string         v_list        = 
         
       "" 
       ;
                    foreach         (PanGu.WordInfo wordInfo         in 
        words)               
            {                   
                        v_list=wordInfo.Word+"^"+wordInfo.Rank+"^"+wordInfo.Frequency+"^"+wordInfo.WordType+"^"+wordInfo.Pos;        
                        //        词        + 
       " 
       ^ 
       " 
       + 
       权重 
       + 
       " 
       ^ 
       " 
       + 
       词频 
       + 
       " 
       ^ 
       " 
       + 
       " 
       ^ 
       " 
       + 
       词性;
                        if        (i        == 
       0 
       ) keyWord  
       = 
        v_list;                 
                        else         keyWord         += 
        separator 
       + 
       v_list;       
                i        ++        ;               
            }               
                    return         keyWord;          
        }       
    }   
}

 

 

运行default.aspx，结果是这个样子滴，呵呵，词^权重^词频^词性，我的程序没有做任何校验和判断，看官要自己加上，要不id缺失，或者表的description为Null，应该会出错滴

问题关键不在这，呵呵，反正有了“词^权重^词频^词性”这个结果，后面的事情不就好办了，切分一下，爱筛选也行，爱按权重词频排序也行，那就不细说了。我计划的应用是自动提取文章的关键词，填入到db_table的keywords字段里面去，这样无论搜索还是输入到页面做SEO，不都挺好用的嘛，呵呵