Part 1 Introducing language models

So what are language models, exactly? And how can we use them for data analysis? This part of the book answers both those questions.

In chapter 1, we discuss the principles underlying language models and what makes them special. We also discuss all the different ways in which language models can be used for data analysis, covering options to use them directly on data as well as the possibility of using them as interfaces to more specialized data-analysis tools.

In chapter 2, we have a “chat” with ChatGPT: that is, we interact with a popular language model by OpenAI. We witness the flexibility of ChatGPT when performing a variety of tasks on text, ranging from text classification to extracting specific pieces of information from text based on a concise task description. We also see that ChatGPT does well when translating questions about data, formulated in natural language, to formal query languages such as SQL.

After reading this part, you should have a good understanding of what language models are and how you can use them for data analysis.