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

5 Crafting a data story using the DIKW pyramid

 

This chapter covers

  • A homelessness tale, our scenario for this chapter
  • What a data story is and how it relates to the DIKW pyramid
  • How to incorporate generative AI into the DIKW pyramid

Creating stories is the most exciting part of data storytelling. It’s really about following a plot with a beginning, a main point, and a conclusion—just like in the stories at the cinema or theater. Here, however, there are constraints imposed by the data. In this chapter, we will focus on using the DIKW pyramid by describing two examples. The first example will analyze the problem of homelessness in Italy, and the second will examine the problem of fake news on a hypothetical website. We will also describe the concept of a data story and how it relates to the DIKW pyramid already discussed in the previous chapters. Finally, we will describe some strategies to incorporate generative AI in the DIKW pyramid, based on how to write specific prompts for ChatGPT to generate the context of a chart and for DALL-E to generate contextual images to incorporate into the chart.

5.1 Breaking the ice: The homelessness tale

5.1.1 What was wrong with the chart?

5.1.2 What was wrong with the presentation?

5.2 Uncovering the narrative: What a data story is

5.2.1 Using the DIKW pyramid to streamline a data story

5.2.2 DIKW in action: Completing the homelessness tale

5.3 Incorporating generative AI into the DIKW pyramid

5.4 Behind the scenes: The homelessness tale

5.4.1 Creating a compelling subtitle

5.4.2 Generating images

5.5 Another example: Fake news

5.5.1 From data to information

5.5.2 From information to knowledge

5.5.3 From knowledge to wisdom

Summary

References