APPENDIX A

The Binary Classification Trade-Off

A binary classification trade-off arises when we wish to classify a set of things into two categories (call them In and Out), but we do not have a direct way of doing the classifying. On the other hand, there is a proxy for those things that is relatively easy to classify. The problem is that the proxy is only approximate. Because it is only approximate, there are four classification outcomes:

Image True positive: The proxy classifies things as In that should be In.

Image True negative: The proxy classifies things ...

Get Principles of Computer System Design now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.