Introduction to Classification: Naïve Bayes and Nearest Neighbour

Abstract

This chapter introduces classification, one of the most common data mining tasks. Two classification algorithms are described in detail: the Naïve Bayes algorithm, which uses probability theory to find the most likely of the possible classifications, and Nearest Neighbour classification, which estimates the classification of an unseen instance using the classification of the instances `closest' to it. These two methods generally assume that all the attributes are categorical and continuous, respectively.