Data Storytelling with Altair and AI cover
welcome to this free extract from
an online version of the Manning book.
to read more
or

7 From information to knowledge: Building textual context

 

This chapter covers

  • Introducing context
  • Calibrating the story to the audience
  • Using ChatGPT for commentaries and annotations
  • Using large language models for textual context
  • A case study: From information to knowledge (part 1)

Talking about knowledge in a computer science book might seem completely out of place. The word knowledge could inspire philosophical concepts or even intimidate. But in this chapter (and the next), we will not be talking about philosophical knowledge but, rather, the knowledge that helps the reader understand the context of a story. It is, therefore, knowledge applied to the context of our data story, rather than general knowledge. In these chapters, we will review the basic concepts behind context in a data story and how to adapt it based on the audience. First, we will focus on textual context in this chapter, and in the next one, we will cover images. We will introduce large language models (LLMs) and use ChatGPT as an example of LLM implementation for data storytelling. Finally, we will explore a practical example.

7.1 Introducing context

7.2 Calibrating the story to the audience

7.2.1 General public

7.2.2 Executives

7.2.3 Professionals

7.3 Using ChatGPT for commentaries and annotations

7.3.1 Describing the topic

7.3.2 Describing the type

7.3.3 Setting custom instructions

7.4 Using large language models for context

7.4.1 Fine-tuning

7.4.2 Retrieval augmented generation

7.5 Case study: From information to knowledge (part 1)

7.5.1 Tailoring the chart to the audience

7.5.2 Using RAG to add a commentary

7.5.3 Highlighting the period of decrease in sales

7.5.4 Exercise

Summary

References

Embeddings

Fine-tuning