Azure Cognitive Services Text Analytics API returns inconsistent sentiment scores for the same input
I'm prototyping a solution and I'm using the Azure Cognitive Services Text Analytics API (v3.1) to analyze sentiment from user feedback in my application. However, I've noticed that the sentiment scores returned for the same piece of text can vary significantly depending on the timing of the request. For example, when I send the following text: ```plaintext "I love using this product! It has greatly improved my workflow." ``` I sometimes receive a score of 0.9 (positive) and at other times, a score of 0.3 (neutral). I have confirmed that the text being sent is exactly the same, and the API key and endpoint remain unchanged. I initially thought this might be due to network latency or issues on Azure's side, so I implemented retry logic with exponential backoff. However, the inconsistencies still persist. I'm using the following code snippet to call the API: ```csharp using Azure.AI.TextAnalytics; using Azure; var client = new TextAnalyticsClient(new Uri("https://<your-resource-name>.cognitiveservices.azure.com/"), new AzureKeyCredential("<your-api-key>")); var response = await client.AnalyzeSentimentAsync("I love using this product! It has greatly improved my workflow."); var sentiment = response.Value; ``` I've verified that I'm not hitting any limits on the number of requests, and my subscription tier is supposed to allow for a high volume of transactions. Has anyone else encountered this issue? What could be causing such unpredictable sentiment analysis results from the API? Any insights or troubleshooting tips would be appreciated! This is part of a larger mobile app I'm building. This issue appeared after updating to C# latest.