Python Feature Engineering Cookbook - Third Edition

4

Performing Variable Discretization

Discretization is the process of transforming continuous variables into discrete features by creating a set of contiguous intervals, also called bins, which span the range of the variable values. Subsequently, these intervals are treated as categorical data.

Many machine learning models, such as decision trees and Naïve Bayes, work better with discrete attributes. In fact, decision tree-based models make decisions based on discrete partitions over the attributes. During induction, a decision tree evaluates all possible feature values to find the best cut-point. Therefore, the more values the feature has, the longer the induction time of the tree is. In this sense, discretization can reduce the time it takes ...