k-Nearest Neighbor (k-NN)

Question 1

2 Type of k-Nearest Neighbor

Answer 1

k-Nearest Neighbor Classifiers
k-Nearest Neighbor Regressors

Question 2

4 Requirement of k-Nearest Neighbour Classifiers

Answer 2

A set of labelled records
Proximity metric - E.g. Euclidean
The value of k - The number of nearest neighbours to retrieve.
A method to determine the class label of unknown record

Question 3

2 Phase of K-NN Algorithm

Answer 3

Training Phase
Store the collected data.

Testing Phase (Prediction)
Calculate distance
Sort and select the nearest point based on k.
Predict the class for the point by majority vote

Question 4

2 Type of Voting Approach

Answer 4

Majority Voting Approach
Distance-Weighted Voting Approach