diff --git a/Data Transformation/codechef_problem/transform_description.py b/Data Transformation/codechef_problem/transform_description.py
index 88055fc..938a702 100644
--- a/Data Transformation/codechef_problem/transform_description.py	
+++ b/Data Transformation/codechef_problem/transform_description.py	
@@ -1,11 +1,17 @@
 __author__ = 'Pranay'
 import lemmatization, minimize_desc, remove_nondecodable_chars, removePunctuation
-
+from nltk.corpus import stopwords
+from nltk.tokenize import word_tokenize
+ 
 def transform(desc):
     print 'Description Transformation Started: '
     desc = remove_nondecodable_chars.removeNondecodableChars(desc)
     desc = minimize_desc.minimizeDescription(desc)
     desc = removePunctuation.removePunctuation(desc)
     desc = lemmatization.lemmatizeDescription(desc)
+    stop_words = set(stopwords.words('english'))
+    word_tokens = word_tokenize(desc)
+    filtered_sentence = [w for w in word_tokens if not w.lower() in stop_words]
+    desc=' '.join(filtered_sentence)
     # print 'Description Transformation Ended '
     return desc
diff --git a/Model Training/get_accuracy.py b/Model Training/get_accuracy.py
index 684b125..b7f2eb0 100644
--- a/Model Training/get_accuracy.py	
+++ b/Model Training/get_accuracy.py	
@@ -96,7 +96,7 @@ def get_recommendations(username, prev_sub, no_recomm, no_test):
 
     with open('accuracy_val.csv', 'w') as f:
         writer = csv.writer(f)
-        writer.writerow(['prev_sub', 'no_recomm', 'no_test', 'tp', 'tn', 'fp', 'fn', 'precision', 'recall', 'f1_score'])
+        writer.writerow(['prev_sub', 'no_recomm', 'no_test', 'tp', 'tn', 'fp', 'fn', 'precision', 'recall', 'f1_score','Specificity','False_Positive_Rate','True_Negative_Rate','False_Negative_Rate'])
 
         prev_sub = 5
         no_recomm = 10
@@ -117,10 +117,13 @@ def get_recommendations(username, prev_sub, no_recomm, no_test):
                     precision = tp/(tp + fp)
                     recall = tp/(tp + fn)
                     f1_score = 2 * precision * recall / (precision + recall)
-
+                    Specificity = tn / tn + fp
+                    False_Positive_Rate = fp / fp + tn
+                    True_Negative_Rate = tn / tn + fp
+                    False_Negative_Rate = fn / fn + tp
                     print "Precision - " + str(precision)
                     print "Recall - " + str(recall)
                     print "F1score - " + str(f1_score)
 
-                    dat = [prev_sub, no_recomm, no_test, tp, tn, fp, fn, precision, recall, f1_score]
+                    dat = [prev_sub, no_recomm, no_test, tp, tn, fp, fn, precision, recall, f1_score,Specificity,False_Positive_Rate,True_Negative_Rate,False_Negative_Rate]
                     writer.writerow(dat)