Supervised Learning - Classification Week 8 Challenge

Q2. Answer

Pros:- 1) it is simple machine learning alogorithm and easy to understand.

2) it acn be used classification and regression .

3) it can be used for multiclass problems.

4) its provide variety od distance criteria to choose from Euclidean distance ,hamming distance ,manhattan distance and minkawski distance.

Cons- 1) its difficult to choose the optimal nomber of neighbors to be consider while classifying new data entry.

2) K-NN inherently has no capability of dealing with missing value problem.

3)K-NN has calculate proximity berween each neighbors for every single value in data sets. for large datasets computational cost is increases