Sentiment analysis is a powerful tool for understanding the emotional tone of text data. The prebuilt sentiment analysis model within the Power Platform can detect whether text is positive, negative, or neutral. This capability is invaluable for analyzing various text sources, such as social media posts, customer reviews, or any other text data where understanding sentiment is crucial. The model provides both scores and labels at the sentence and document levels, offering a comprehensive view of the sentiment expressed. While scores can be positive, negative, or neutral, documents can also receive a “mixed” sentiment label when there isn’t a clear predominant emotion, and this label comes without a score. The overall document sentiment is determined by aggregating the individual sentence sentiments.
Exploring Sentiment Analysis in Power Apps
Before integrating the sentiment analysis model into your cloud flows, you can easily explore its capabilities within Power Apps.
- Begin by signing in to either Power Apps or Power Automate.
- In the left-hand pane, navigate to … More and then select AI hub.
- Choose AI models.
- Locate and select the Sentiment analysis – Detect positive, negative, or neutral sentiment in text data model.
- You can then analyze predefined text samples or input your own text. Click Analyze text to see how the model evaluates your input.
For those looking to embed these AI Builder sentiment analysis models directly into Power Apps Studio, the formula bar offers integration capabilities. You can find more detailed instructions in the guide on Using Power Fx in AI Builder models in Power Apps (preview).
Leveraging Sentiment Analysis in Power Automate
If your workflow requires the use of this prebuilt model within Power Automate, you can find comprehensive guidance in the article titled Using the sentiment analysis prebuilt model in Power Automate.
Supported Languages and Data Format
The sentiment analysis model supports a wide range of languages, including German, Spanish, English, French, Hindi, Italian, Japanese, Korean, Dutch, Norwegian, Portuguese (Brazil), Portuguese (Portugal), Turkish, Chinese (Simplified), and Chinese (Traditional). It’s important to note that documents processed by the model cannot exceed 5,120 characters in length.
Model Output Explained
When text is processed, the sentiment analysis model provides the following key pieces of information:
- Sentiment: This can be categorized as Positive, Negative, Neutral, or Mixed.
- Confidence score: A numerical value ranging from 0 to 1. Scores closer to 1 indicate a higher degree of confidence in the accuracy of the detected sentiment.
- Sentences: A detailed breakdown of individual sentences from the input text, each with its own sentiment analysis.
- Sentiment: Each sentence is classified as Positive, Negative, Neutral, or Mixed.
- Sentence confidence score: Similar to the document-level score, this value ranges from 0 to 1, with higher values indicating greater confidence in the sentence’s sentiment accuracy.
Understanding Usage Limits
The following limits apply to calls made per environment across several prebuilt models, including language detection, sentiment analysis, and key phrase extraction:
| Action | Limit | Renewal Period |
|---|---|---|
| Calls (per environment) | 400 | 60 seconds |
This ensures efficient and controlled usage of the AI capabilities within your Power Platform environment. For additional related information, please refer to the Related information section.