You are using KNN for a regression problem. What are the special considerations in selecting K and the distance metric, and how would you evaluate the model's performance?

  • Choose K and metric considering data characteristics, evaluate using regression metrics
  • Choose fixed K and Manhattan metric, evaluate using recall
  • Choose large K and any metric, evaluate using accuracy
  • Choose small K and Euclidean metric, evaluate using precision
Selecting K and distance metric considering the data characteristics and evaluating the model using regression metrics like RMSE or MAE is the right approach for KNN in regression.
Add your answer
Loading...

Leave a comment

Your email address will not be published. Required fields are marked *