Amazon Q in QuickSight is a natural language querying feature in Amazon QuickSight BI tools that allows users to ask questions about their data in plain language and receive visual responses instantly. This feature helps users create visual diagrams and analyze datasets with the assistance of a large language model (LLM) in the background.
To start using QuickSight, you must first sign up via the QuickSight AWS Console. You can log in to QuickSight using SAML, an IAM user, or IAM Identity Center.
Setting up an IAM Role for QuickSight
You need to allow access to the resources you plan to use with QuickSight. For this example, I used a dataset stored in an S3 bucket.
Once the user is successfully created, you can proceed by uploading your dataset to S3, which serves as the repository. QuickSight supports many different dataset types as sources.
After successfully uploading the dataset, it becomes visible in QuickSight. For this demonstration, I used a dataset related to the sales of home materials at offline retail stores.
My user successfully created.
Now we put our dataset in the S3 as a repository. Quicksight support many different datasets as a source.
After dataset upload successfully, we can see our data in the Quicksight.
I used data set about sales of home material at retail offline stores.
Optimizing the Dataset in QuickSight
In this stage, you can create synonyms and remove unused labels to optimize the dataset for both the QuickSight dashboard and QuickSight Q (Generative AI for BI). This step is particularly useful when enabling the generative AI capabilities later.
QuickSight User Roles and Licenses
To create a dashboard from scratch and enable generative AI features in QuickSight, you need an AUTHOR PRO license.
Using Generative AI in QuickSight
With generative AI in QuickSight, you can issue prompts to generate visualizations based on your source dataset using the "Build a Visual" command. This feature allows you to ask questions about your dataset, create diagrams, and gain deeper insights for analysis.
Top comments (0)