In this chapter, we first looked at the details of neural networks. We started by looking at how neural networks have evolved over the years. We studied different types of neural networks. Then, we looked at the various building blocks of neural networks. We studied in depth the gradient descent algorithm, which is used to train neural networks. We discussed various activation functions and studied the applications of activation functions in a neural network. We also looked at the concept of transfer learning. Finally, we looked at a practical example of how a neural network can be used to train a machine learning model that can be deployed to flag forged or fraudulent documents. 

Looking ahead, in the next chapter, we will look into how we can use such algorithms for natural language processing. We will also introduce the concept of web embedding and will look into the use of recurrent networks for natural language processing. Finally, we will also look into how to implement sentiment analysis.