I have a 4-leabelled text classification problem.

Could someone help me choose among the below text classifiers ?

I was advised to select the second one ( the one which uses both unigrams and bigrams ) but I cannot really see why.

enter image description here

  • 1
    $\begingroup$ What is the problem statement? What is being predicted here? $\endgroup$ – BlackCurrant May 23 '20 at 17:28
  • $\begingroup$ I have built two text classifiers for a reduced version ( 4-labels only instead of 20) of the scikit-learn.org/stable/modules/generated/… dataset.. Above you can see the accuracy, precision, classification report and confusion matrix of the two learned classifiers $\endgroup$ – batman May 24 '20 at 9:54

Okay so keeping it very short and precisely in context of your question-

Accuracy tells us, out of all the documents how many are classified correctly.

Precision tells us out of all documents which are predicted in a category, how often its correct.

Uni -gram- "nasa", "is" "space" , "agency" bi-gram- "nasa is", "space agency"

Now lets go over the numbers, in both the cases accuracy and precision doesn't have significant difference.

But as we can see bi-grams can give me much more information and hence can have better performance on unseen data. Try to test the model on unseen data/validation set and compare the difference.May be Try tri-grams etc also.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.