Let’s download the Netflix Engagement Report from the first half of 2023 and upload the Excel file into ChatGPT to see what we can learn.
(Feel free to also open the file in Tableau to investigate and create visualizations.)
Section Generative AI and Data Analytics and Visualization
In Section we talked a little about generative AI in the context of creating and describing data visualizations.
As of July 2024, there are many examples of AI chatbots that are commonly used. A few examples are:
- ChatGPT
- Google Bard
- Microsoft Copilot
- Jasper
The field of generative AI is constantly changing, and so by the time you read what is written here, it will almost certainly already be outdated.
At the moment, there are many things that generative AI is useful for, and there are things that it cannot do (reliably) quite yet. It is important that we know how to use AI in our work but also understand its current limitations. It is still important for us to have a good understanding of what we are doing so that we are not presenting misleading visualizations or analysis.
For example, here is a YouTube video showing how ChatGPT-4 did not provide correct analysis of a relatively simple data set:
Newer versions of ChatGPT allow us to upload a data file, and the chatbot can use Python behind the scenes to analyze the data. We’re going to play around more with what it can currently do in terms of analyzing data and/or creating visualizations.
