Data Science For Dummies, 2nd Edition

Chapter 7

Modeling with Instances

IN THIS CHAPTER

Overviewing instance-based learning

Distinguishing classification from cluster methods

Working through average nearest neighbor algorithms

Mastering the core features and characteristics of k-nearest neighbor algorithms

Seeing how nearest neighbor algorithms can be used in the retail sector

Data scientists use classification methods to help them build predictive models that they can then use to forecast the classification of future observations. Classification is a form of supervised machine learning: The classification algorithm learns from labeled data. Data labels make it easier for your models to make decisions based on the logic rules you’ve defined. Your plain-vanilla clustering algorithm, like the k-means method, can help you predict subgroups from within unlabeled datasets. But there’s more to life than plain vanilla. I think it’s about time to take things one step further, by exploring the instance-based family of machine ...