ClassificationKNN

K-nearest neighbor classification model.

📝Syntax

mdl = fitcknn(X, Y)
mdl = fitcknn(X, Y, Name, Value)
label = predict(mdl, Xnew)
[label, score, cost] = predict(mdl, Xnew)

📥Input Arguments

Parameter	Description
X	numeric matrix: rows are observations and columns are predictors.
Y	vector: class labels with one label for each row of X.
Name, Value	optional name-value arguments accepted by the corresponding fitting function.

📤Output Arguments

Parameter	Description
mdl	classification model object returned by the fitting function.
label	predicted class labels for new observations.
score	class scores or posterior-like values when the model provides them.

📄Description

ClassificationKNN stores a nearest-neighbor classifier with its training predictors, class labels, distance metric, and neighbor count.

Create this object with fitcknn. Use predict to classify observations from their nearest neighbors.

💡Examples

Train a nearest-neighbor classifier and classify two observations.

X = [0 0; 0 1; 1 0; 5 5; 5 6; 6 5];
Y = [1; 1; 1; 2; 2; 2];
mdl = fitcknn(X, Y, 'NumNeighbors', 3);
label = predict(mdl, [0.2 0.1; 5.2 5.1])

🔗See Also

Used Functions

fitcknn predict

🕔Version History

Version	Description
2.0.0	initial version