#!/usr/local/bin/perl5
&main_loop;

##########################################################
##  Details of this program:
##  The program is divided into the following steps and can be underst
+ood as follows:
##  1.The location of files are initialized to scalar variables for ea
+sy location of the files themselves wherever needed
##  2.The subroutine init_vector is called with an argument(namely: pl
+ant,train,place/person).It will for each of the given argument
##    by the user store the weight as a "value"of an array of hash whe
+re the key value is the "term"and the index is the serial no of the
##    .I document i.e the document.Also the subroutine will be designe
+d in such a way that by just changing the fixed weight value in the 
##    subroutine we can easily implement either of the two proposed we
+ighing schemes of part 2.
##  3.The subroutine for calculation of centroids is defined. 
##  4.Then there will be subroutines for calculation of full cosine si
+milarity which will take the inputs from  the two centroid  vectors 
##    and return their similarity values,the differences in their valu
+es and hence the sorted array.This is then printed out here
##    itself. 
##     Also printed out in the end is the ratio of correct /total coun
+ts.
##     
##
## 
##  
########################################################## 



##########################################################
##  INIT_INPUT_FILENAMES,
##
##  This function specifies the names and locations of
##  input files used by the program. 
##
##  Parameter:  $type   ("stemmed" or "unstemmed")
##
##  If $type == "stemmed", the filenames are initialized
##  to the versions stemmed , while
##  in the default ("unstemmed") case initializes to files
##  containing raw, unstemmed tokens.
##  
##  
##########################################################

sub init_input_filenames{ 
  local($type) = @_;
 
  $DIR=".";
  
  if ($type eq "stemmed") {
    $plant               ="$DIR/plant.stemmed";
    $plant_hist          ="$DIR/plant.stemmed.hist";
    $plant_titles        ="$DIR/plant.titles";
    $tank                ="$DIR/tank.stemmed";
    $tank_hist           ="$DIR/tank.stemmed.hist";
    $tank_titles         ="$DIR/tank.titles";
    $perplace            ="$DIR/perplace.stemmed";
    $perplace_hist       ="$DIR/perplace.stemmed.hist";
    $perplace_titles     ="$DIR/perplace.titles";
    $common_words        ="$DIR/common_words.stemmed";

  } else {
    $plant               ="$DIR/plant.tokenized";
    $plant_hist          ="$DIR/plant.tokenized.hist";
    $plant_titles        ="$DIR/plant.titles";
    $tank                ="$DIR/tank.tokenized";
    $tank_hist           ="$DIR/tank.tokenized.hist";
    $tank_titles         ="$DIR/tank.titles";
    $perplace            ="$DIR/perplace.tokenized";
    $perplace_hist       ="$DIR/perplace.tokenized.hist";
    $perplace_titles     ="$DIR/perplace.titles";
    $common_words        ="$DIR/common_words";

  }
}

sub main_loop{
    print "Please enter the choice for using the stemmed(1) or unstemm
+ed documents(2)\n enter choice:";
    $opt_st = <STDIN>;
    if ($opt_st !~/[12]/){$opt_st = 1;}

    chomp($opt_st);
    if($opt_st == 1){
    &init_input_filenames("stemmed");
    }else{
    &init_input_filenames("tokenized");
    }
        
    
    print "Please enter the choice for the parameter for which the sim
+ilarity values would be calculated:plant(1),tank(2),person/place(3)\n
+ enter choice:\n opt1:";
    $opt1 = <STDIN>;
    chomp($opt1); 
    if ($opt1 !~/[123]/){$opt1 = 1;}
    print $opt1;
    
    
    print "Please enter a number for the options you want to have for 
+weighing : normal(1),eponential decay(2),stepped (3)\n";  
    print "enter choice for opt2:";    
    $opt2 = <STDIN>;
    print"I am a aperl monk";
    
    chomp($opt2);      
    if($opt2 !~ /[123]/){$opt2 = 1;}
    
    print $opt2;
[download]