Summary

This chapter was aimed at learning the current trends in NLP and learning the future directions that NLP is being driven to. Though it is a very broad topic, we discussed some of the very recent advancements that have been made in NLP. As current trends, we first looked at the advancements being made with regard to word embeddings. We saw that much more accurate embeddings with richer interpretations (for example, probabilistic) are emerging. Then we looked into improvements that have been made in machine translation, as it is one of the most sought after areas in NLP. We saw that better attention mechanisms and better MT models capable of producing increasingly more realistic translations are both emerging.

We then looked at some of the novel research in NLP that is taking place (mostly in 2017). First we investigated the penetration of NLP into other fields: computer vision, reinforcement learning, and the generative adversarial models. We looked at how NLP systems are being improved...