AI Builder Actions in Power Automate are a powerful set of tools that allow users to leverage Artificial Intelligence (AI) to automate tasks and processes. AI Builder Actions enable users to easily create custom models to enable automated tasks such as object recognition, form processing, language understanding, text recognition, and more.
AI Builder Actions enable users to automatically classify and extract text from images, forms, and documents. This allows users to quickly process large amounts of data and extract the necessary information without manual input. AI Builder Actions can also be used to recognize objects in images and videos, allowing users to quickly identify items or people in the picture.
AI Builder Actions also enable users to build custom models that can be used to detect sentiment from text, allowing users to better understand customer sentiment and take the appropriate action. AI Builder Actions can also be used to identify intent from text, enabling users to better understand customer needs and respond accordingly.
Overall, AI Builder Actions are a powerful set of tools that allow users to quickly and easily build custom models to automate tasks and processes. AI Builder Actions empower users to leverage AI to automate mundane tasks and processes, freeing up time and resources to focus on more important tasks.
In this series of blogs we will discuss some of the AI Builder Actions –
Analyze positive or negative sentiment
Power Automate can be used to detect sentiment in natural language, such as customer emails, social media posts, and website feedback. It can also be used to detect sentiment in unstructured data, such as images or videos. The sentiment analysis feature can be used to quickly identify customer sentiment and take the appropriate action.
Overall, Power Automate makes it easy to analyze positive or negative sentiment in text. With AI Builder Actions, users can create custom models to accurately detect sentiment in text and take the appropriate action. This helps businesses better understand customer sentiment and take the necessary steps to improve customer service.
Inputs:
- Language: This is a dropdown that allows you to select the language. Currently, it supports Chinese, Dutch, English, French, German, Hindi, Italian, Japanese, Korean, Norwegian, Portugese, Spanish and Turkish.
- Text: This is the text to analyze.
Outputs:
| Name | Type | Description | Values |
| Probability overall text is negative | float | Probability of the negative sentiment in the analyzed text | Value in the range of 0 to 1. Values close to 1 indicate greater confidence that the identified sentiment is accurate |
| Probability overall text is neutral | float | Probability of the neutral sentiment in the analyzed text | Value in the range of 0 to 1. Values close to 1 indicate greater confidence that the identified sentiment is accurate |
| Probability overall text is positive | float | Probability of the positive sentiment in the analyzed text | Value in the range of 0 to 1. Values close to 1 indicate greater confidence that the identified sentiment is accurate |
| sentences | List | List of sentence data structures containing sentences overall sentiment and scores | Sentence sentiment, positive, neutral and negative scores |
| Overall text sentiment | string | Overall sentiment of the analyzed text | Positive, neutral or negative |
