Create strengths_and_weaknesses.txt

strengths_and_weaknesses.txt

@@ -0,0 +1,9 @@
+strengths
+- the model is trained with almost 15.000 comments and the distribution of the rates are balanced
+- it can predict the rating very accurate if the actual rating is in the interval [1,4]
+- exclamation mark and emojis effects the output rating
+
+weaknesses
+- the model is having trouble with predicting the rating if the actual rating is close to 0 and 5
+- longer comments tends to be considered as average rating since it increases the neutral score
+- exclamation mark and emojis effects the output rating