Extracting Features from Text Variables

Text can be part of the variables in our datasets. For example, in insurance, information about an incident may come from free text fields in a form. On a website that gathers customer reviews, some information may come from short text descriptions provided by the users. Text data does not show the tabular pattern of the datasets we have worked with throughout this book. Instead, information in texts can vary in length and content, and the writing style may be different. We can still extract a lot of information from text variables to use as predictive features in machine learning models. The techniques we will cover in this chapter belong to the realm of Natural Language Processing (NLP). NLP is a subfield of linguistics and computer science, concerned with the interactions between computer and human language, or, in other words, how to program computers to understand human language. NLP includes a multitude of techniques to understand the syntax...