Sentiment Analysis
Category: Cognitive
Description: Advanced extension module utilizing Azure sentiment analysis, offering sentiment labels such as "negative," "neutral," and "positive" determined from the highest confidence score detected by the service, both at the sentence and document level. Additionally, this feature provides confidence scores ranging from 0 to 1 for each sentence and document, reflecting the positive, neutral, and negative sentiments present within them.
Functionality: This module allows understanding and analyzing text data from clients, identifying expressed emotions and attitudes. With this information, opinions about products and services can be evaluated, customer satisfaction can be enhanced during conversations with agents or bots, and content and recommendations can be personalized.
For more detailed information, access Here
Prerequisites
-
Create Resource in Azure
- Sign in to portal.azure.com.
- Navigate to "Azure AI services | Language Service" and select "Create resource"; this resource should be of type TextAnalytics.
For a guide to create a language resource, access Here
Implementation
Expression to evaluate: Field of type string that allows entering an expression to evaluate. It can be completed with an entity or text which must be entered within single quotes to be identified as such.
Language of the text to evaluate: List containing languages to select for analysis of entered text.
Entity where the evaluation result will be deposited: Field where you must select or create an entity in which to deposit the API result.
Sentiment Analysis Subscription Key: Enter the key for Azure Cognitive Services TextAnalytics found in the Keys and Endpoint section. Azure provides two options: key1 and key2, either can be used.
Sentiment Analysis Endpoint: Enter the key for Azure Cognitive Services TextAnalytics found in the Keys and Endpoint section.
Intention to execute in case of error. The action flow will be interrupted, and the error description will be passed to the selected intention named 'AnalyzeSentiment_Error': Dropdown list where the name of the intent for error handling must be selected.
Implementation Example
| Field | Value |
|---|---|
| Name: | SentimentAnalysisTest |
| Expression to evaluate: | 'Hello I am very angry because my purchase did not arrive' |
| Language of the text to evaluate: | en |
| Entity where the evaluation result will be deposited: | testEntity |
| Sentiment Analysis Subscription Key: | '0XX6XX5XX7edXXXXXXXXXXXXXXXXXXXX' |
| Sentiment Analysis Endpoint: | 'https://mynewresource.cognitiveservices.azure.com/' |
| Intention to execute in case of error. The action flow will be interrupted, and the error description will be passed to the selected intention named 'AnalyzeSentiment_Error': | Error_Handler |