In this “Sentiment Analysis Using Power Automate AI Builder With Examples” article, we will learn how to work with sentiment analysis using Power Automate with examples.
Power Automate, Microsoft’s cloud-based service for automating workflows, continues to evolve with powerful AI capabilities. One such feature is the “Analyze positive or negative sentiment in text” action, which allows users to incorporate sentiment analysis into their automated workflows. This capability is particularly useful for businesses seeking to understand customer feedback, monitor social media sentiments, or enhance overall user experience. In this article, we’ll delve into how to use this AI builder action in Power Automate, accompanied by real-world examples.
Sentiment Analysis Using Power Automate AI Builder: Understanding Sentiment Analysis
Sentiment analysis, also known as opinion mining, involves determining the emotional tone or attitude expressed in a piece of text. The “Analyse positive or negative sentiment in text” action utilises pre-trained models to assign a sentiment score to text inputs. The sentiment score ranges from 0 to 1, where 0 indicates a highly negative sentiment, 1 indicates a highly positive sentiment, and 0.5 suggests a neutral sentiment. This automation way helps organisations with customer satisfaction survey reports, where they get to know how their customer service is. If any negative feedback comes from customers, they can improve their services for better business continuity and growth.
Use cases of Sentiment Analysis Using Power Automate AI Builder
There could be many use cases where we can implement sentiment analysis using Power Automate; a few are given below:
Example 1: Customer Feedback Analysis
Consider a scenario where a company receives customer feedback through an online form. Using Power Automate, you can automatically analyze the sentiment of the feedback and take different actions based on the result. For instance:
If the sentiment score is below 0.5, create a task for customer support to address the issue.
If the sentiment score is above 0.5, send a thank-you email and add the feedback to a positive testimonials database.
Example 2: Social Media Monitoring
In a social media monitoring scenario, you can use Power Automate to analyse the sentiment of tweets mentioning your brand. If the sentiment is positive, automatically retweet or respond with appreciation. If the sentiment is negative, create a task for the social media team to address the issue promptly.
Example 3: Employee Satisfaction Survey
For HR workflows, you can integrate sentiment analysis into employee satisfaction surveys. Analyzing open-ended responses can provide insights into the overall sentiment of the workforce. Based on sentiment scores, you can trigger follow-up actions like organizing team-building activities for a positive sentiment or addressing concerns for a negative sentiment.
Setting up the “Analyze positive or negative sentiment in text” Action:
- Open Power Automate and create a new flow or open an existing one.
- Add a trigger or action that provides text input, such as “When a new email arrives” or “When a new tweet is posted.”
- Add the “Analyze positive or negative sentiment in text” action from the AI Builder category.
- Configure the action by providing the text input from the previous step.
Let’s see the above steps practically.
Demo Examples: Sentiment Analysis Using Power Automate AI Builder
The typical automated example of sentiment analysis is “When an email arrives in Outlook,” read that email sentiment and create a task in the SharePoint Online list if the sentiment is negative for further action. However, for this proof of concept, I will create a manually triggered instant cloud flow.
Create a manually triggered flow and add an text type input.
Click on the “Add an action” icon.
Search this keyword: “AI Builder”
AI Builder in Power Automate
From the list of AI Builder actions, select the “Analyze positive or negative sentiment in text” action.
The configuration of the “Analyse positive or negative sentiment in text” action is very simple; we just need to pass the language and text parameters. Here I have selected language as English and dynamic text input from the previous trigger action steps (manually trigger flow). You can pass the static text as input as well.
Finally, we added a compose action to capture the result from the previous step, “Analyse positive or negative sentiment in text.”.
That’s it. Now my complete flow looks like below:
This looks very simple, isn’t it?
Now, let’s test this flow out.
I have passed the below text in the input box when I clicked on the test flow link:
“I recently purchased a new laptop, and I am extremely satisfied with its performance. The speed and graphics quality exceeded my expectations. The customer service was also excellent, addressing my queries promptly. Overall, I highly recommend this brand to others.”
Click on the “Run flow” button.
Now, let’s expand the output of the compose data operation. There, we can also see the same sentiments as shown below:
Few Other Texts for the Power Automate Sentiment Analysis Test
Text 1:
“I encountered several issues with the software I purchased. The user interface is confusing, and the customer support was unhelpful in resolving my concerns. I am frustrated and disappointed with the overall experience. I expected better quality for the price I paid.”
Text 2:
“Just wanted to express my gratitude for the outstanding service I received from your team. The support agent went above and beyond to assist me with my technical issues. It’s rare to find such dedication and expertise in customer service. Kudos to your team!”
Text 3:
“I’m having trouble with the latest update of your app. The new features are not user-friendly, and the app frequently crashes. This has negatively impacted my productivity, and I hope these issues can be resolved soon. Disappointed with the current performance.”
Text 4:
“Attended your virtual event last week, and I must say it was a fantastic experience. The speakers were engaging, and the content was highly informative. The platform was user-friendly, making it easy to navigate between sessions. Looking forward to future events!”
We can use this text as input text in the “Analyse positive or negative sentiment in text” action to observe how the sentiment analysis scores vary for different tones and expressions.
Summary: Sentiment Analysis using Power Automate AI Builder
Thus, in this article, we have learned about how to work with the “Analyze positive or negative sentiment in text” action in Power Automate AI Builder to analyse the user input about the organisation’s service. By incorporating sentiment analysis into your Power Automate flows, you can automate responses, prioritise tasks, and gain valuable insights from unstructured text data. As businesses strive for more customer-centric approaches, leveraging AI in tools like Power Automate becomes a strategic advantage for enhancing the user experience and making data-driven decisions.
See Also: Power Platform Articles
You may also visit the Power Platform article hub, where you will see a bunch of articles focusing on Power Platform, like Power Automate, Power Apps, etc. All the articles are written with real-time project scenarios and troubleshooting techniques.
If you found this article helpful and enjoyed it, please consider sharing it with your friends and colleagues. Please don’t forget to subscribe to our site to receive our latest articles directly in your inbox. 🙂