This chapter closes the explanation of the NLP implementation process with Scala. In this chapter and the previous one, we evaluated different frameworks for this programming language, and the pros and cons of each have been detailed. In this chapter, the focus has been mostly on a DL approach to NLP. For that, some Python alternatives have been presented, and the potential integration of those Python models in a JVM context with the DL4J framework has been highlighted. At this stage, a reader should be able to accurately evaluate what will be the best fit for his/her particular NLP use case.
Starting from the next chapter, we will learn more about convolution and how CNNs apply to image recognition problems. Image recognition will be explained by presenting different implementations using different frameworks, including DL4J, Keras, and TensorFlow.
