AI-102 Question 94
Select 3You are building a content moderation system for a social media platform using Azure AI Content Safety. The platform must automatically detect and flag offensive language in user-generated text while also identifying potential hate speech or sexually explicit content. Which steps should you take to implement this solution?
- A
Enable the Azure AI Content Safety service and configure the text moderation API to analyze user-generated content.
- B
Train a custom Azure Machine Learning model to classify offensive language and integrate it with Azure AI Content Safety.
- C
Define a severity level threshold in Azure AI Content Safety to determine when content should be flagged.
- D
Use the Azure AI Content Safety API to extract sentiment scores for identifying hate speech.
- E
Integrate the Azure AI Content Safety API with the platform's backend to process and flag content in real-time.
Show answer and explanation
Correct answers: A, C, E
Explanation
To implement a text moderation solution using Azure AI Content Safety, you need to enable and configure the service, define severity levels for flagging content, and integrate the API with your system for real-time processing. Azure AI Content Safety provides built-in models for detecting offensive language, hate speech, and explicit content, so training a custom model is unnecessary. Sentiment analysis is not part of Azure AI Content Safety's functionality for this use case.
- A. Correct.
Correct: Enabling the Azure AI Content Safety service and configuring the text moderation API is a fundamental step to utilize its capabilities for analyzing user-generated content.
- B. Incorrect.
Incorrect: While custom models can be trained using Azure Machine Learning, the Azure AI Content Safety service already provides pre-built models for detecting offensive language, hate speech, and explicit content. Training a custom model is unnecessary in this scenario.
- C. Correct.
Correct: Setting a severity level threshold in Azure AI Content Safety allows you to define when content should be flagged based on its level of offensiveness or risk.
- D. Incorrect.
Incorrect: The sentiment analysis feature is not used to identify hate speech in Azure AI Content Safety. Instead, the service uses specialized moderation models to detect such content.
- E. Correct.
Correct: Integrating the Azure AI Content Safety API with the platform's backend ensures that content is processed and flagged in real-time, meeting the system's requirements.