Exploring watsonx.data through a sample project

This topic provides a steps to explore the unstructured data ingestion, enrichment and retrieval capabilities of watsonx.data as a Service using a sample data set packaged into a sample project.

Before you begin

  • Provision IBM watsonx.data as a Service instance. For information about provisioning, see Setting up watsonx.data as a Service.

  • Create Mulvus connection.

  • Create Presto connection with a schema in iceberg catalog.

Procedure

  1. After provisioning watsonx.data, the watsonx.data context home page is displayed with a welcome message. Click Take a tour to navigate through the watsonx.data interface.

    watsonx.data Premium tile

  2. After completing the tour, click Get started with sample project.

    watsonx.data Premium tile

  3. You will be prompted to create an API key.

    a. Click the User settings link in the Missing user API key message.

    watsonx.data Premium tile

    b. Click User API Key > Create a key.

    watsonx.data Premium tile

    watsonx.data Premium tile

    Note: It is not required to copy or download the API key.
  4. After the API key is created, click watsonx on the top left corner to return to the home page. -->

  1. On the Sample project page,

    a. The sample project name is pre-populated. You can use the default name or change it if required. You can specify the optional fields as required.

    b. Select the Storage.

    Important:

    A default storage is created when you create an IBM cloud account. If you don’t have the default storage, click Cloud Object Storage and follow the instructions to create a COS bucket.

    c. You can use the toggle switches to enable or disable the project controls in the Advanced settings section.

    d. Select the watsonx.ai Runtime instance.

    e. Verify the values and click Create. A progress bar is displayed.

    watsonx.data Premium tile

    watsonx.data Premium tile

    f. After the sample project is created, click View new project. The sample project page opens.

    watsonx.data Premium tile

  1. From the Assets tab, click New asset+ and select Analyze and enrich unstructured data. The Unstructured data curation page opens. This starts the unstructured data curation flow.

  2. Create an unstructured data curation asset and save the details. The Discover your document landscape page opens.

  3. Click Next. You can view the Filter your document for processing details.

  4. Click Next. To process the documents for extraction, click Get started.

  5. Click Add documents. The documents can reside as an asset inside your project or in your computer or in a connected source such as Cloud Object Storage or Amazon S3 connection.

Documents in your Project

To place your documents in your Project, you have the following two ways:

**Option1**: To add your documents to your project through the Upload data files option. To do that, see [Adding data to a project](../manage-data/add-data-project.html).

**Option2**: If your documents reside in any connected sources, you can import the files to your project, To do that, see [Convert files in project storage to assets](../manage-data/add-data-project.html).

Documents in a connected source

To place your documents in a connected source such as Cloud Object Storage or Amazon S3 connection, see [Manage connections](../manage-data/conn_types.html).

For example, you can directly choose documents from the following supported connected source -

Documents in your local system

You can directly choose and upload a local file to be used.
  1. The documents uploaded get listed. Click Analyze documents.

  2. After the document analysis is complete, you can proceed to process the document. Click Process documents. The Process documents page opens.

  3. Wait for the success message. Click Go to document library from the notification window or click View assets button to navigate to document library overview page.

You can also access the document library overview page from your project > Assets > Data > Document libraries and select the invoice-document-library. This is the document library that was created as part of the unstructured data curation flow.

watsonx.data Premium tile

  1. In the document library overview page, click Chat in prompt lab.

    watsonx.data Premium tile

  2. You can take the tour or skip the tour. In the chat interface, start asking questions to get insights into your data.

watsonx.data Premium tile

watsonx.data Premium tile

The following are some sample prompts:

• What is the total spend across invoices?

• Sort invoices spend in an ascending order

• Sort invoices spend in a descending order

• Calculate the average invoice total

• Who is the seller on the invoice with the highest amount?

• Give me Ship to details from all the invoices