Read and Review Automation Anywhere Documentation

Automation 360

Close Contents


Open Contents

Microsoft LUIS NLP package

  • Updated: 2021/09/02
    • Automation 360 v.x
    • Build
    • RPA Workspace

Microsoft LUIS NLP package

The Microsoft LUIS NLP package contains actions that enable you to connect to and consume the Microsoft Cognitive Services Text Analytics API to identify the language, sentiment, key phrases, and entities. This package supports the following languages: English, Chinese (Simplified), French, German, and Spanish.

Important: This is a beta package and is currently not available with the Automation 360 Enterprise and Cloud editions.

Before you start

You must have an Azure Cognitive Services resource to use these actions. See Create a Cognitive Services resource using the Azure portal.

You also require the following in order to use the actions:

  • Service Endpoint URL: the endpoint URL that identifies the Azure service.
  • Subscription Key: the unique key that authenticates Automation Anywhere Enterprise.

Actions in the Microsoft LUIS NLP package

Action Description
Detect language Identifies the language of the provided content and returns it in ISO 639-1 language code. The output is stored in a string variable.
Get key phrases Identifies the main points and returns a list of key phrases. For example, if the input text is about a basketball game, this action returns the names of teams, the name of the venue, and the final score.
Get named entities Identifies the entities in the provided content such as people, places, organizations, date/time, quantities, branded products, and book titles.

The output is stored in a dictionary variable, where each name is a key, and the corresponding entity is the value.

Get sentiment Analyzes the provided content and returns a sentiment and score.
  • If the score is greater than 0.6, the sentiment is Positive.
  • If the score is in the 0.2 through 0.6 range, the sentiment is Neutral.
  • If the score is less than 0.2, the sentiment is Negative.

The output is stored in a dictionary variable containing two keys and their corresponding values: sentiment and score.

Send Feedback