October 2020
Beginner to intermediate
448 pages
15h 6m
English
In the last chapter, we discussed the basics of clustering algorithms, which are exploratory algorithms without defined outcomes or labels. In this chapter, we’ll explore predictive algorithms, which are algorithms that use historical data to predict new outcomes. Predictive algorithms have defined outcomes or labels; they are also called supervised learning algorithms. These algorithms are particularly useful when trying to predict user behavior or distributional needs.
This chapter examines some predictive models that should be in every analyst’s toolkit:
k-Nearest neighbor
Ordinary least squares (OLS)
Logistic regression
Decision trees
Support vector machines ...
Read now
Unlock full access