diff --git a/scripts/dataset/ground_truth_arrange.py b/scripts/dataset/ground_truth_arrange.py
index 17b39c4..47e2dd3 100644
--- a/scripts/dataset/ground_truth_arrange.py
+++ b/scripts/dataset/ground_truth_arrange.py
@@ -1,94 +1,107 @@
 # Copyright (C) 2015-2016  The Software Heritage developers
 # See the AUTHORS file at the top-level directory of this distribution
 # License: GNU General Public License version 3, or any later version
 # See top-level LICENSE file for more information
 
 #!/usr/bin/env python3
 # coding: utf-8
 
 import os
 import sys
 import io
 import json
-from shutil import copyfile
 
 def main(root):
     
     root_ground_truth = root + '/../ground_truth'
     root_ground_truth_text = root_ground_truth + '_text'
     root_code_by_language = root + '/../code_by_language'
     counts = dict()
-
+    
     try:
         os.mkdir(root_code_by_language)
     except FileExistsError:
         pass
-        
+    
     for r in os.listdir(root):
         if not r.startswith('.'):
             for d in os.listdir(root + '/' + r):
                 if not d.startswith('.'):
                     try:
-                        ground_truth = io.open(root_ground_truth + '/' + r + '/' + d)
+                        ground_truth = io.open(root_ground_truth + '/' + r + '/' + d + '.json')
                         try:
                             j = json.load(ground_truth)
-                            for k in j.keys():
-                                root_language = root_code_by_language + '/' + k
+                            for language in j.keys():
+                                root_language = root_code_by_language + '/' + language
                                 try:
                                     os.mkdir(root_language)
                                 except FileExistsError:
                                     pass
-                                for f in j.get(k):
-                                    counts[k] = counts.get(k, 0) + 1
+                                for f in j.get(language):
+                                    counts[language] = counts.get(language, 0) + 1
+                                    start = (counts[language] - 1) // 1000 * 1000 + 1
+                                    end = start + 999
+                                    root_count = root_language + '/' + str(start) + '-' + str(end)
+                                    if counts[language] % 1000 == 1:
+                                        try: 
+                                            os.mkdir(root_count)
+                                        except FileExistsError:
+                                            pass
                                     (_,ext) = os.path.splitext(f)
-                                    new_name = str(counts[k]) + ext
+                                    new_name = str(counts[language]) + ext
                                     copy_src = root + '/' + r + '/' + d + '/' + f
-                                    copy_des = root_language + '/' + new_name
-                                    copyfile(copy_src, copy_des)
-                                    print(copy_des + ' successfully copied.')
+                                    copy_des = root_count + '/' + new_name
+                                    os.symlink(copy_src, copy_des)
+                                    print(copy_src + ' successfully copied.')
                         except json.decoder.JSONDecodeError:
                             ground_truth.close()
-                            ground_truth = io.open(root_ground_truth_text + '/' + r + '/' + d, 'r', encoding="utf-8")
+                            ground_truth = io.open(root_ground_truth_text + '/' + r + '/' + d, 'r')
                             while(True):
                                 line = ground_truth.readline()
                                 if line == '\n' or line == '':
                                     break
                                 else:
-                                    stripped = line.strip()
-                                    print(stripped)
                                     pass
-                        
+                                
                             while(True):
                                 line = ground_truth.readline()
                                 stripped = line.strip()
                                 if line == '':
                                     break
                                 else:
                                     stripped = line.strip()
                                     language = stripped.replace(':','')
                                     root_language = root_code_by_language + '/' + language
                                     try:
                                         os.mkdir(root_language)
                                     except FileExistsError:
                                         pass
                                     while(True):
                                         line = ground_truth.readline()
                                         if line == '\n':
                                             break
                                         else:
                                             f = line.strip()
                                             counts[language] = counts.get(language, 0) + 1
+                                            start = (counts[language] - 1) // 1000 * 1000 + 1
+                                            end = start + 999
+                                            root_count = root_language + '/' + str(start) + '-' + str(end)
+                                            if counts[language] % 1000 == 1:
+                                                try: 
+                                                    os.mkdir(root_count)
+                                                except FileExistsError:
+                                                    pass
                                             (_,ext) = os.path.splitext(f)
                                             new_name = str(counts[language]) + ext
                                             copy_src = root + '/' + r + '/' + d + '/' + f
-                                            copy_des = root_language + '/' + new_name
-                                            copyfile(copy_src, copy_des)
-                                            print('{} successfully copied.'.format(copy_des))
+                                            copy_des = root_count + '/' + new_name
+                                            os.symlink(copy_src, copy_des)
+                                            print('{} successfully copied.'.format(copy_src))
                     finally:
                         ground_truth.close()
-
-if __name__ == "__main__":
+                            
+if __name__ == '__main__':
     if len(sys.argv) != 2:
-        print("Only argument acceptable is a path.")
+        print('Only argument acceptable is a path.')
     else:
         main(sys.argv[1])
diff --git a/scripts/dataset/repo_arrange_text.sh b/scripts/dataset/repo_arrange_text.sh
index dd5057d..dd39810 100755
--- a/scripts/dataset/repo_arrange_text.sh
+++ b/scripts/dataset/repo_arrange_text.sh
@@ -1,29 +1,29 @@
 #!/bin/bash
 
 ROOT_FOLDER=$1
 OLDIFS=$IFS
 IFS=$'\n'
 
 if [ $# -eq 1 ]
 then
     mkdir $ROOT_FOLDER/../ground_truth_text/;
     for root in $( ls $ROOT_FOLDER -1 );
     do
 	for repo in $( ls "$ROOT_FOLDER/$root" -1 );
 	do
-            GROUND_TRUTH=$ROOT_FOLDER/../ground_truth/$root/$repo;
+            GROUND_TRUTH=$ROOT_FOLDER/../ground_truth/$root/$repo.json;
 	    if [[ -s $GROUND_TRUTH ]];
 	    then
 		echo "Ground truth of '$repo' exist!";
 	    else
 		mkdir "$ROOT_FOLDER/../ground_truth_text/$root"
-		linguist "$ROOT_FOLDER/$repo" --breakdown > $ROOT_FOLDER/../ground_truth_text/$root/$repo;
+		linguist "$ROOT_FOLDER/$root/$repo" --breakdown | grep -ax '.*' > $ROOT_FOLDER/../ground_truth_text/$root/$repo;
 		echo "Done";
 	    fi
 	done
     done
 else
     echo "Please enter root folder correctly. One folder is needed."
 fi
 
 IFS=$OLDIFS