Chapter 1. What is machine learning?
Figure 1.1. Machine-learning process for the cats and dogs competition
Figure 1.2. The loan-approval process for the microlending example
Figure 1.3. After a few months of business and 2,500 loan applications, 1,000 were approved, of which 700 applicants repaid the loan on time and the other 300 defaulted. This initial set of observed information is critical to start building automation into your loan-approval process.
Figure 1.4. Filtering new applications through two business rules enables you to reduce manual analysis to only 52% of the incoming applications.
Figure 1.5. Basic ML workflow, as applied to the microloan example
Figure 1.6. In this two-class classification, individual data points can belong to either the round class or the square class. This particular data lies in a two-dimensional feature space having a nonlinear decision boundary that separates the classes, denoted by the curve. Whereas a simple statistical model does quite poorly at accurately classifying the data (center), an ML model (right) is able to discover the true class boundary with little effort.
Figure 1.7. The workflow of real-world machine-learning systems. From historical input data you can build a model using an ML algorithm. You then need to evaluate the performance of the model, and optimize accuracy and scalability to fit your requirements. With the final model, you can make predictions on new data.