Hands-On Data Science for Marketing

Logistic regression versus decision trees

If you recall from the previous chapter, a logistic regression model learns from the data by finding the linear combination of the feature variables that best estimates the log odds of an event occurring. Decision trees, as the name suggests, learn from the data by growing a tree. We are going to discuss how decision tree models grow and to build trees in more detail in the following section, but the main difference between the logistic regression and decision tree models is the fact that logistic regression algorithms search for a single best linear boundary in the feature set, whereas the decision tree algorithm partitions the data to find the subgroups of data that have high likelihoods of an event ...