How can KNN classify if there are more than 2 dimension

Is it possible to use the similar example from Matlab to classify with 4 dimensions.

Or does this example only classify according to 2 dimensions? load fisheriris X = meas; Y = species; Mdl = fitcknn(X,Y,'NumNeighbors',4); %% % Predict the classification of an average flower. flwr = mean(X); % an average flower flwrClass = predict(Mdl,flwr)

Answer 1

[Akira Agata]

Yes, it is possible to use fintcknn to classify with 4 dimensional data. The following code (same as your example) can generate classification model using all 4 dimensions, and predict an average flower.

load fisheriris
X = meas;
Y = species; 
Mdl = fitcknn(X,Y,'NumNeighbors',4);
flwr = mean(X);
flwrClass = predict(Mdl,flwr);

[Selina Loh]

Can we plot to see the distance of the nearest neighbours?

Thank you

[Akira Agata]

What do you mean by 'plot' for 4 dimensional data? If you want to find the nearest neighbor, the following can do that.

D = pdist2(flwr,X);
[~,idx] = min(D);

Then, idx was found to be 65. Thant means the nearest neighbor from flwr in dataset X is X(65,:) ('versicolor').

[Selina Loh]

Thank you :)