Practical Natural Language Processing

Chapter 7. Topics in Brief

The problems are solved, not by giving new information, but by arranging what we have known since long.

Ludwig Wittgenstein, Philosophical Investigations

So far in Part II of this book, we’ve discussed a few common application scenarios of NLP: text classification, information extraction, and chatbots (Chapters 4 through 6). While these are the most common use cases for NLP we’re likely to encounter in industry projects, there are many other NLP tasks that are relevant in building real-world applications involving large collections of documents. We’ll take a quick look at some of these topics in this chapter. Let’s first start with a few largely unrelated scenarios you may encounter in your workplace projects. We’ll discuss them in more detail throughout the chapter.

If someone asks us to find out what NLP is and we have no idea, where do we start? In the pre-internet era, we would’ve hit the nearest library to do some research. However, now the first place we’d go is to a search engine. Search involves a lot of human-computer interaction using natural language, so it gives rise to very interesting use cases for NLP.

Our client is a big law firm. When a new case comes up, they sometimes have to research lots and lots of documents related to the case to get a bigger picture of what it’s about. Many times, there isn’t enough time for a thorough manual review. Our client wants us to develop software that can provide a quick overview of the topics discussed ...