Extract Key Phrases
Description
The Extract Key Phrases activity identifies the most important phrases and concepts within text using Azure AI Language services.
Key phrases help summarize the main topics discussed in the text and can be used for categorization, indexing, search, and downstream automation.
For example, from the text:
The customer reported issues with invoice processing and payment reconciliation.
The activity may extract:
- invoice processing
- payment reconciliation
- customer
Common settings
Use Process Data to bind supported inputs to workflow values or upstream outputs. See Common Activity Features for shared execution, error-handling, and output-persistence settings. Availability depends on the activity.
Design-Time Configuration
Display Text
The label for this activity as it appears on the workflow canvas.
Connection Name
Select one of the following:
- An Azure Text Analytics connection configured in AI Command Center using customer-provided Azure credentials.
- IB-X Azure Text Analytics Credits to use the IntelliBuddies Integration Gateway.
When using IB-X Azure Text Analytics Credits, requests are processed through the Integration Gateway and consume IB-X Currency based on the request volume and service usage.
Input Text
The text from which key phrases should be extracted.
The value can be entered directly or supplied dynamically from Process Data.
Execution settings
Configure Wait All Incoming when execution paths converge, and On Error for failure handling. Select Inherit where available to use the workflow-level error policy.
Outputs
Key Phrases
A collection of key phrases extracted from the supplied text.
Raw Response
The complete JSON response returned by Azure AI Language services.
Notes
- Key phrases represent the most significant concepts identified in the text.
- The activity may return multiple phrases for a single text input.
- Extraction quality generally improves with longer and more descriptive text.
- When using the Integration Gateway, IB-X Currency is consumed based on the request volume.
Example: Analyze text
Select an Azure AI Language connection or the documented credits option, then bind Input Text to text such as “The customer needs help with invoice processing.”. After execution, use Key Phrases to inspect each returned phrase.
For a collection output, check it contains items before iterating. Keep Raw Response for cases where the workflow needs details not exposed by the structured output. See Azure Text Analytics connection.