This chapter covers
- Defining products for human-in-the-loop machine learning applications
- Creating a system for exploratory data analysis for short text
- Creating an information extraction system to support a human process
- Creating an image labeling system to maximize model accuracy
- Evaluating options for extending simple systems
This final chapter contains three worked examples of human-in-the-loop machine learning products. Using everything that you have learned in the first 11 chapters, you will implement three examples. You can think of these examples—one for exploratory data analysis of news headlines, one for extracting information about food safety from text, and one for labeling images containing bicycles—as being first-pass systems that you can create in a few days. The examples are similar to the human-in-the-loop machine learning system in chapter 2 but slightly more sophisticated, building on what you have learned in the chapters since.