Contents

Description of demo_binaryclass_decisions.m

Demonstrates stump, tree, and forest binary classification on three different datasets

clear all
close all
generateData_vert

usage of stump binary classification (vert data)

options_st = [];
model_st = ml_binaryclass_stump(Xtrain, ytrain, options_st);
yhat_st = model_st.predict(model_st, Xtest);
testError_st = mean(yhat_st ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_st.name, testError_st);

Averaged misclassification test error with Binary Decision Stump is: 0.036

usage of tree binary classification (vert data)

options_tr = [];
options_tr.dataTypes = [1 1];
options_tr.maxDepth = 16;
model_tr = ml_binaryclass_tree(Xtrain, ytrain, options_tr);
yhat_tr = model_tr.predict(model_tr, Xtest);
testError_tr = mean(yhat_tr ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_tr.name, testError_tr);

Averaged misclassification test error with Binary Decision Tree is: 0.418

usage of random forest binary classification (vert data)

options_rf = [];
options_rf.oobMaxFeatures = [1 2];
model_rf = ml_binaryclass_randomForest(Xtrain, ytrain, options_rf);
yhat_rf = model_rf.predict(model_rf, Xtest);
testError_rf = mean(yhat_rf ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_rf.name, testError_rf);

Averaged misclassification test error with Random Forest Binary Classification is: 0.236

figure;
plot2DClassifier(Xtrain, ytrain, model_st);
figure;
plot2DClassifier(Xtrain, ytrain, model_tr);
figure;
plot2DClassifier(Xtrain, ytrain, model_rf);

generateData_slanted

usage of stump binary classification (slanted data)

options_st = [];
model_st = ml_binaryclass_stump(Xtrain, ytrain, options_st);
yhat_st = model_st.predict(model_st, Xtest);
testError_st = mean(abs(yhat_st - ytest));
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_st.name, testError_st);

Averaged misclassification test error with Binary Decision Stump is: 0.498

usage of tree binary classification (slanted data)

options_tr = [];
options_tr.dataTypes = [1 1];
options_tr.maxDepth = 16;
model_tr = ml_binaryclass_tree(Xtrain, ytrain, options_tr);
yhat_tr = model_tr.predict(model_tr, Xtest);
testError_tr = mean(yhat_tr ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_tr.name, testError_tr);

Averaged misclassification test error with Binary Decision Tree is: 0.258

usage of random forest binary classification (slanted data)

options_rf = [];
options_rf.oobMaxFeatures = [1 2];
model_rf = ml_binaryclass_randomForest(Xtrain, ytrain, options_rf);
yhat_rf = model_rf.predict(model_rf, Xtest);
testError_rf = mean(yhat_rf ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_rf.name, testError_rf);

Averaged misclassification test error with Random Forest Binary Classification is: 0.329

figure;
plot2DClassifier(Xtrain, ytrain, model_st);
figure;
plot2DClassifier(Xtrain, ytrain, model_tr);
figure;
plot2DClassifier(Xtrain, ytrain, model_rf);

generateData_groups

usage of stump binary classification (groups data)

options_st = [];
model_st = ml_binaryclass_stump(Xtrain, ytrain, options_st);
yhat_st = model_st.predict(model_st, Xtest);
testError_st = mean(yhat_st ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_st.name, testError_st);

Averaged misclassification test error with Binary Decision Stump is: 0.498

usage of tree binary classification (groups data)

options_tr = [];
options_tr.dataTypes = [1 1];
model_tr = ml_binaryclass_tree(Xtrain, ytrain, options_tr);
yhat_tr = model_tr.predict(model_tr, Xtest);
testError_tr = mean(yhat_tr ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_tr.name, testError_tr);

Averaged misclassification test error with Binary Decision Tree is: 0.476

usage of random forest binary classification (groups data)

options_rf = [];
options_rf.oobMaxFeatures = [1 2];
model_rf = ml_binaryclass_randomForest(Xtrain, ytrain, options_rf);
yhat_rf = model_rf.predict(model_rf, Xtest);
testError_rf = mean(yhat_rf ~= ytest);
fprintf('Averaged misclassification test error with %s is: %.3f\n', ...
        model_rf.name, testError_rf);

Averaged misclassification test error with Random Forest Binary Classification is: 0.169

figure;
plot2DClassifier(Xtrain, ytrain, model_st);
figure;
plot2DClassifier(Xtrain, ytrain, model_tr);
figure;
plot2DClassifier(Xtrain, ytrain, model_rf);

Published with MATLAB® R2015a