In section 3 you will learn the basics and terminology of classification. You will use histograms to build a “simple” classifier and learn about an important concept called the Bayes error. Then we will move on to making classifiers using nearest means and nearest neighbours. Finally, we discuss how to use the nearest neighbour classifier in practice. After this section you can:

• • Apply the histogram classifier, KNN and nearest mean with pen and paper
  • Recognize pros and cons of a histogram classifier
  • Explain the effect of the hyperparameter K for KNN
  • Compute the error rate and accuracy of a classifier
  • Explain the concept of Bayes error and class overlap