azure-samples / cognitive-search-templates

This project contains samples including Power BI template file for visualizing AI enrichments in the Azure Search knowledge store

page_type	languages	products	name	description	urlFragment
sample	Power BI Template	azure azure-search	Power BI template for knowledge store	A Power BI template with visualizations intended for knowledge stores created using the Import data wizard in the Azure portal.	/powerbi/templates

Power BI sample template for knowledge store (Azure Search)

This repository contains a Power BI template that provides visualizations for a knowledge store created by the Import data wizard in the Azure portal.

In the Import data wizard, next to the template download link, look for and save the following values:

• Skillset name
• Content field
• Enrichment granularity
• Storage account

These values are required by the sample template.

Contents

This repository contains the following items.

File/folder	Description
powerbi/	Folder containing the sample Power BI template.
.gitignore	Define what to ignore at commit time.
CONTRIBUTING.md	Guidelines for contributing to this project.
SECURITY.md	Instructions for reporting security bugs.
README.md	This README file.
LICENSE	The license for the sample.

Prerequisites

• Power BI Desktop

• A knowledge store created in the Azure Portal with the Import data wizard. To create a knowledge store, in the Import data wizard, on the Add cognitive search page (second page in the wizard), fill out all sections including:

  • Add enrichments: These options create the skillset based on the skills you select.
  • Save enrichments to a knowledge store: These options create the knowledge store. You must select Azure table projections to save the enriched content to tables. This wizard section also contains a link to this repository and shows values that will be required to initialize the template: skillset name, content field, enrichment granularity, and storage account. Copy and save these values.

  After you complete the Import data wizard, wait for the knowledge store to be populated with the enriched content before you open the sample template. To verify that the knowledge store content is available, use the Storage Explorer in the Azure portal to view the table projections in the Azure storage account.

Download the sample template

In this repository, go to the cognitive-search-templates folder and click Clone or Download to clone the repository or download a ZIP file. If you download a ZIP file, right-click the file to extract its contents.

Connect the Power BI template with your knowledge store

1. Start Power BI Desktop.

2. Go to File > Import > Power BI Template, and select the cognitive-search-content-analytics-template.pbit file.

3. When prompted, provide the following parameters:

   • Skillset name
   • Source data field (content field): This is the field you chose to enrich. It is often named content or merged_content.
   • Enrichment granularity level (document, pages, or sentences)
   • Knowledge store storage account name

   These are the values you saved during the Import data wizard process.

   On first use, you may also be prompted for the storage account access key.

Analyze your data with Power BI visualizations

Once you have the data loaded from Azure tables into Power BI, you can explore the latent relationships, hierarchy, and structure of your enriched content.

Key concepts

This is an example of a Power BI template. It includes visualizations that tend to work well with enriched content. You can use this example as-is, or adapt it to accommodate more complex skillsets or projections.

cognitive-search-templates's Issues

Azure

Adtivae

Activattion

1. 1
2. 2
3. 3
4. 4
5. 5

Azure

Error using template

Hello, I receive this error after specifying connection parameters:

Feedback Type:
Frown (Error)

Error Message:
The database compatibility level of 1465 is below the minimal compatability level of 1480 needed for [table 'Document'].[partition 'Document-00520621-884d-4a0e-b737-eddffb05e15c' in table 'Document']::[QueryGroup] at PowerBI mode.