Chapter 2. Classifying with k-Nearest Neighbors

 

This chapter covers

  • The k-Nearest Neighbors classification algorithm
  • Parsing and importing data from a text file
  • Creating scatter plots with Matplotlib
  • Normalizing numeric values

Have you ever seen movies categorized into genres? What defines these genres, and who says which movie goes into what genre? The movies in one genre are similar but based on what? I’m sure if you asked the people involved with making the movies, they wouldn’t say that their movie is just like someone else’s movie, but in some way you know they’re similar. What makes an action movie similar to another action movie and dissimilar to a romance movie? Do people kiss in action movies, and do people kick in romance movies? Yes, but there’s probably more kissing in romance movies and more kicking in action movies. Perhaps if you measured kisses, kicks, and other things per movie, you could automatically figure out what genre a movie belongs to. I’ll use movies to explain some of the concepts of k-Nearest Neighbors; then we will move on to other applications.

2.1. Classifying with distance measurements

2.2. Example: improving matches from a dating site with kNN

2.3. Example: a handwriting recognition system

2.4. Summary