The term deep in deep learning refers to a neural network model having several layers, and the learning takes place with the help of data. And based on the type of data used, deep learning may be categorized into two major categories, as shown in the following screenshot:

As shown in the preceding diagram, the type of data used for developing a deep neural network model can be of a structured or unstructured type. In Chapter 2, Deep Neural Networks for Multi-Class Classification, we illustrate the use of a deep learning network for classification problems using structured data where the response variable is of the categorical type. In Chapter 3, Deep Neural Networks for Regression, we illustrate the use of a deep learning network for regression problems using structured data where the response is a continuous type of variable. Chapters 4 to 12 illustrate the use of deep learning networks for mainly two types of unstructured data that involve images and text. In chapters 4 to 8, we provide application examples of some popular deep learning networks using image data, which is regarded as an unstructured type of data. Finally, in chapters 9 to 12, we cover some popular deep learning networks that are useful with text data, which is another major category within unstructured data.

Now, let's briefly go over the examples and techniques covered in chapters 2 to 12.