Thread: text categorization

Hi,

I am trying to start the following project:

A system that can look through text documents and be able to classify the document into 1 of 5 categories, similar to a newspaper, which places different articles in different sections like Sports, Finance, etc. Has anyone ever done something like this before? Can anyone advise on how to implement something like this in C? I appreciate any ideas.

Thanks.

I would make arrays with terms for that area Ex:

Code:

char sports[4][]={"Saints","VooDoo","Hornets","LSU"};

Of course it would all be lowecase and then you would open the file tolower each word and compare it with all the arrays if match occurs categorize.

thanks for the reply. what's the best to way to go through each word in the file and compare it? also, would you be familiar as to how i could dynamically add new keywords to each category? for example, for each document that is added to the category i would like to add new keywords for that category which i obtain from that file.

thanks.

to get a single word from a file and compare would be something like this

Code:

#include <stdio.h>
#include <stdlib.h> /*for exit*/
#include <string.h>

char *getword(FILE *ffile);

int main(void){
     FILE *myfile;
     char example[4][]={"Saints","VooDoo","Hornets",LSU"};
     char array[BUFSIZ];
     int x=0;

     myfile=fopen("temp","rb");
     if(!myfile){
          fprintf(stderr,"Error");
          exit(0);
     }
     array=getword(myfile);   /*gets one word(you would have to put this in a loop*/
     while(x<4){
          if(!strcmp("EOF",array)){
                printf("End of File\n");
                exit(0);
          }
          else if(!strcmp(example[x],array)){
               printf("There is the word!!!!!!\n");
               x++;
               }
          else{
               x++;
          }
}

char *getword(FILE *ffile){
      char buffer[BUFSIZ];
      char word[BUFSIZ];
      char endoffile[]={"EOF"};
      int y=0,int z=0;
      if((fgets(buffer,sizeof buffer,ffile))!=NULL){
           while(buffer[y]!=' ' || (ispunct(buffer[y])!=0) || buffer[y]!='\t' || buffer[y]!='\n'){
            buffer[y]=word[y];
            y++;
            }
       return word;
      }
      else
           return endoffile;
}

I think that will work for a word. But once again I can't check