Publication date: 10/01/2019

K Nearest Neighbors

Predict Response Values Using Nearby Observations

The K Nearest Neighbors platform is available only in JMP Pro.

The K Nearest Neighbors platform predicts a response value for a given observation using the responses of the observations in that observation’s local neighborhood. It can be used for classification of a categorical response as well as for prediction of a continuous response.

K Nearest Neighbors is a nonparametric method that is based on the distance to neighboring observations. Because of this fact, K Nearest Neighbors is able to classify observations using irregular predictor value boundaries. However, because the algorithm is sensitive to irrelevant predictors, the selection of predictors can impact your results.

K Nearest Neighbors has been used successfully in many applications, such as classifying satellite imagery and electrocardiogram (EKG) patterns.

Figure 7.1 Example of the K Nearest Neighbors Platform 

Contents

Overview of the K Nearest Neighbors Platform

Example of K Nearest Neighbors with Categorical Response

Example of K Nearest Neighbors with Continuous Response

Launch the K Nearest Neighbors Platform

The K Nearest Neighbors Report

Continuous Responses
Categorical Responses

K Nearest Neighbors Platform Options

Model Fit Options
Want more information? Have questions? Get answers in the JMP User Community.