Arrange the following steps in a k-nearest neighbors analysis of a classification problem in the most appropriate order (assume k is fixed):
a) For each point to be classified, find the k closest training observations and classify using the most common class.
b) For each point to be classified, calculate the distance from each training observation using an appropriate distance metric.
c) Standardize the predictor/feature variables if appropriate to do so.
What order do a, b, c go in for a k-nearest neighbors analysis
Recently Asked Questions
- What are some good “remote” causes of why someone would want to attend college? In general
- How should I work on a problem that utilizes logs I’m having a little difficulty trying to solve equations that use logs
- Why each organisation should have policies, procedures and practices that support effective risk identification ?