0
$\begingroup$

Please tell me if you see this paper in the link below has used weighted KNN? because they have used weights as the training and testing samples and no formula written. They don't explain the procedure of their KNN. Have they used Euclidean distance for that?

Paper link: https://we.tl/t-H0vKm0g9iD

| improve this question | |
$\endgroup$
0
$\begingroup$

The reference for their KNN algorithm is Zhang et al. 2017, furthermore in their paper they reference that they are using the standard settings for the toolkit (k = 5 etc). In my understanding the weighting they are doing is extracting features out of the text not within the actual KNN algorithm. If their KNN algorithim was weighted then the paper would have specified 'weighted KNN'

| improve this answer | |
$\endgroup$
  • $\begingroup$ Thank you so much, your answer was so helpful. I still have some inquiry because I am a master student and every step should be justified in the thesis. if I wanted to use weighted KNN here, should I have to use the weights resulted from the features extraction? I how can I justify the difference between these Weights of feature extraction and the weights I get using the inverse/sum of inverses for the weighted KNN? $\endgroup$ – Mona Aug 30 '19 at 22:49

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.