Retrieve and process sentiment

Creating a Cognitive Services Resource

To begin, you need to create a Cognitive Services resource in the Azure portal. Follow these steps:

1. Sign in to the Azure portal (https://portal.azure.com/).
2. Click on “Create a resource” in the top-left corner of the portal.
3. Search for “Cognitive Services” and select it from the available resource options.
4. Click on “Create” to start the creation process.
5. Provide the required information such as subscription, resource group, and region.
6. Choose a unique name for your resource and select the “Text Analytics” service.
7. Configure the additional settings and pricing tier according to your needs.
8. Review and create the resource.

Once your Cognitive Services resource is created, you can retrieve the necessary credentials, including the API key, to access the Text Analytics API.

Retrieving Sentiment Using the Text Analytics API

Now, let’s see how to retrieve sentiment using the Text Analytics API with the help of code snippets. First, import the required packages:


import requests


Next, define the API endpoint and key:


api_endpoint = "https://.api.cognitive.microsoft.com/text/analytics/v3.0/sentiment"
api_key = ""


Replace `` with the region of your Cognitive Services resource (e.g., westus) and `` with the API key you obtained for the Text Analytics API.

To retrieve sentiment from a given text, you need to send a POST request to the API endpoint with the text data in the request body. The response will contain the sentiment score ranging from 0 to 1, where 0 indicates a negative sentiment and 1 indicates a positive sentiment.


def analyze_sentiment(text):
headers = {
"Content-Type": "application/json",
"Ocp-Apim-Subscription-Key": api_key
}
data = {
"documents": [
{
"language": "en",
"id": "1",
"text": text
}
]
}
response = requests.post(api_endpoint, headers=headers, json=data)
if response.status_code == 200:
sentiment_score = response.json()["documents"][0]["sentiment"]
return sentiment_score
else:
raise Exception("Error occurred while retrieving sentiment.")


In the `analyze_sentiment` function, the text is passed as a parameter, and a POST request is made to the API endpoint using the `requests` library. The sentiment score is extracted from the response JSON and returned.

You can now use the `analyze_sentiment` function to retrieve sentiment for different pieces of text. Here’s an example:


text = "I am thrilled to work on designing and implementing Microsoft Azure AI solutions!"
sentiment_score = analyze_sentiment(text)
print(f"Sentiment score: {sentiment_score}")


This code will output the sentiment score of the provided text. You can experiment with different texts and analyze their sentiments accordingly.

Conclusion

By leveraging the Text Analytics API from Azure Cognitive Services, you can easily retrieve sentiment related to designing and implementing a Microsoft Azure AI solution. This sentiment analysis can provide valuable insights and help improve the overall experience and effectiveness of your AI solution.

Remember to handle exceptions and error cases properly when working with APIs, and ensure that you have appropriate permission and authentication to access the API endpoint.

In this article, we explored how to retrieve sentiment using Azure Cognitive Services and the Text Analytics API. By incorporating sentiment analysis into your AI solution, you can gain a deeper understanding of user sentiments and make informed decisions based on the results.