Create Topic

With Q, we are introducing a new asset type - Topic. Topic lets you pull in your data sets, enrich it further with user friendly field names, more synonyms, filters, named entities, semantic info and make the data available for Q to answer questions. In this module, we will create a topic with some sample data.

Exercise 1 - Create a dataset
If you already have SaaS-Sales dataset (from Author workshop), skip to Exercise 2.

  1. Download SaaS-Sales.csv
  2. In QuickSight, from Datasets view, click New dataset button.
  3. Select Upload a file option and choose the SaaS-Sales.csv file from local.
  4. Click Edit settings and prepare data button.
  5. Click Save & publish / Save.
  6. Click QuickSight icon to exit dataset edit screen.

Exercise 2 - Create a topic

  1. From Topics view, click the Skip option.
    (It is recommended that you setup the sample topic and explore it as well. For now, let’s skip it to get to creating a new topic.)
  2. Click Skip anyway button in the confirmation dialog.
  3. From top right, click New topic button.
  4. Enter name as SaaS-Sales, Paste in the description - Use this topic to ask questions regarding software sales and click Continue button.
    Your topic users will see this description.
  5. In Add data to topic pop up, select Datasets option, choose SaaS-Sales from dataset list and click Add data button.
  6. Note the banner on top that states that Q has started to index the data in the dataset.

    In next module, we are going to explore what Q can do while indexing and immediately after indexing is complete. There is lot more we can do within the topic to enhance Q’s capabilities. So, please be sure to explore the Enhance Topic module afterwards.