Logo image
Classification: Assigning Observations to Known Categories - Chapter 11
Book chapter

Classification: Assigning Observations to Known Categories - Chapter 11

Jason S Schwarz, Chris Chapman and Elea McDonnell Feit
01 Jan 2020

Abstract

In this chapter, we will explore supervised learning methods. Unlike with clustering, generally, the value of a supervised model output is inherent in the framing of the question. This makes interpretation easier, but it requires an outcome variable to have a strong relationship with its indicator variables, and benefits from data that are well structured and clean. With statistical modeling, people often say “garbage in, garbage out,” meaning that even a very sophisticated model will not be able to produce reliable results if the data are not high quality or there is no actual relationship between input and output variables.

Metrics

23 Record Views

Details

Logo image