praser languege java

Hi everybody:) I am new in java. i try to build the program for phrase-language-lexicon.
this is Tree Builder:
Probabilistic LFG Parser
i need to write same output result in java program. I'v got two text file (lexicon.txt and rules.txt) . lexicon contains words and what those word means in grammar (man-none, bite-verb..etc) in rules file contains rules of position those words in sentience (np-n-v . vp-v ...etc) I need parsing the lexicon file and add it in vector , and parsing entered string from user and add it in vector. after matched above vectors elements for identification words from user-string after match with rules and identification position in sentence . give out print the result ,as in builder tree(s[np[n [man]] vp[v[likes]ad[a]] n[dog]]]. I complicate dump with matching vectors elements....
i bee very appreciate if give me any way how can i do this , or simple examples , o link o same code ...
sorry about my language (english is not my first)

Code Java:

// Read in a String from the user
      //  create a parser to test it against
 
      //   check each word against the lexicon
      //   1. write out the part of speech
 
      // 2. Then check these parts of speech against
      // the rules
      //as in old example open file, read line from lexicon file and create lex
 
 
SOME as:
 
import java.util.StringTokenizer;
import java.util.Vector;
import java.io.BufferedReader;
import java.io.FileReader;
 
public class newtoken {
 
	public static void main(String[] args) throws Exception 
	{
		Vector predlo =new Vector();
		String s = "THE DOG BITES A MAN";
		StringTokenizer st = new StringTokenizer(s, " \t\n\r,.");
          while (st.hasMoreTokens())
	    	 {
             predlo.add(st.nextToken());                                      
		     }
             System.out.println("Thi is a sentence"+predlo);
                FileReader fr = new FileReader("lexicon.txt");
     		    BufferedReader br = new BufferedReader(fr);
                String line4 = br.readLine();
     		while (line4!=null)
     		{
     			//System.out.println(line4);
     			String[] tr = line4.split("\\s");
     			for(int j=0; j<predlo.size(); j++)
                if(predlo.elementAt(j).equals(tr[0]))
                System.out.println(tr[1]+ " "+"["+predlo.elementAt(j)+"]");
 
                line4 = br.readLine();
     		}
 
     	}
	}
 
output :Thi is a sentence[THE, DOG, BITES, A, MAN]
a [A]
v [BITES]
n [MAN]
n [DOG]
d [THE]
 Now need to commpareTo  with Node class for praseOfSpeech...
 
class Node{
 
    String pos;
         // np
 
    Node(String s){
	  pos = s;
 
	}
 
 
 
	////////////////////////////////////////////////////
	//
	////////////////////////////////////////////////////
    private String getNode(){
		   if (!checkPos())
	      getNode();
	      //getout/clause  pass a parameter in the next call
	      // perform a 're-write'
 
	   else
	     return pos;  // only an entry that is in the PartsOfSpeech vector
	                  //  may be returned here
    }
 
    ////////////////////////////////////////////////////
    //
    ////////////////////////////////////////////////////
    public boolean checkPos(){
	   boolean flag = false;
 
	   for(int i =0;i<POS.size();i++)            // Go through entire POS vector
	      if (pos = (String)POS.elementAt(i))    // Check is it there
		      flag = true;                       // if so, then return true
 
	  return flag;
	}
 
}