Try Different Classification Models

Instructions are in the task pane to the left. Complete and submit each task one at a time.

load letterdata.mat

traindata

Task 1

histogram(traindata.Character)

Task 2

boxplot(traindata.MADX,traindata.Character)

Make Predictions with a Classification Model

Instructions are in the task pane to the left. Complete and submit each task one at a time.

This code loads the data and fits a kNN model.

load letterdata.mat

traindata

knnmodel = fitcknn(traindata,"Character","NumNeighbors",5,"Standardize",true,"DistanceWeight","squaredinverse");

testdata

Task 1

predLetter = predict(knnmodel,testdata)

Task 2

misclassrate = sum(predLetter ~= testdata.Character)/numel(predLetter)

Task 3

testloss = loss(knnmodel,testdata)

 

Investigate Misclassifications

Instructions are in the task pane to the left. Complete and submit each task one at a time.

This code loads the data and a trained kNN model and makes a confusion chart.

load letterdata.mat

load predmodel.mat

testdata

predLetter

confusionchart(testdata.Character,predLetter);

Task 1

confusionchart(testdata.Character,predLetter,"RowSummary","row-normalized");

Task 2

falseneg = (testdata.Character == "U") & (predLetter ~= "U");

Task 3

fnfiles = testfiles(falseneg)

fnpred = predLetter(falseneg)

Task 4

badU = readtable(fnfiles(4));

plot(badU.X,badU.Y)

Investigate Features of Misclassified Samples

Instructions are in the task pane to the left. Complete and submit each task one at a time.

This code loads the data and a trained kNN model.

load letterdata.mat

load predmodel.mat

traindata

testdata

predLetter

Task 1

idx = (traindata.Character == "N") | (traindata.Character == "U");

UorN = traindata(idx,:)

idx = (testdata.Character == "U") & (predLetter ~= "U");

fnU = testdata(idx,:)

Task 2

UorN.Character = removecats(UorN.Character);

Task 3

UorNfeat = UorN{:,1:end-1};

fnUfeat = fnU{:,1:end-1};

Task 4

parallelcoords(UorNfeat,"Group",UorN.Character)

Task 5

hold on

plot(fnUfeat(4,:),"k")

hold off