@S18CRX0120
KNN learns no parameters and also not memorize anything ,then why option 1 is correct.help!!!
KNearest Neighbour QUIZ
Hey @Bhawna, knn is memory based approach, this means that knn has to learn all training data, ( do not consider this as learning parameters ). And since it is a memory based approach, whenever a new training data comes, it learns that as well and stores it in its memory. So i is correct as well.
Hope this resolved your doubt.
Plz mark the doubt as resolved in my doubts section. 
@S18CRX0120
But in KNN ,We first calculate distances of querypoint from all points and then we search for k nearest points and then take majority .what are we memorizing here??
By memorizing means r u trying to say that now in future ,knn has memorized class of point for which we have already calculated correct class earlier???
Hey @Bhawna, you are getting it wrong, as you said,
So for calculating distance of query point from all points, you should have coordinates of all those points stored in the memory. Other wise you will not be able to do so.
Since we need to store/memorize all training points that is why it is called memory based approach.
@S18CRX0120
In Linear regression also,We have training data stored in memory will it mean that Linear Regression is also a memory based approach .
Also Time complexity for single query is nlogn and as n increases there is no linear growth here how is option 2 correct???
Hey @Bhawna, yes in linear regression also there is training data, but do you need to remember all training datapoints, while making predictions??
No, we just only need to remember thetas and bias terms to make predictions. Since we need not to cram all data points its not memory based algorithm. Although we need to remember values of thetas and bias, these are parameters, so its a parametric algorithm.
Yes the question needs correction, informed the mentor.
Hope this resolved your doubt.
Plz mark the doubt as resolved in my doubts section.
Understood finally…
I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.
On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.