AI-102 Question 93
Select 2You are designing a text moderation solution for a social media platform using Azure AI Content Safety. The platform requires content to be analyzed for offensive language and hate speech. Additionally, the system should classify the severity of violations to decide whether to block, flag, or allow the content. Which combination of Azure AI Content Safety features should you implement to fulfill these requirements?
- A
Enable text content detection for offensive language and hate speech.
- B
Configure severity levels for detected violations.
- C
Use Azure Translator to translate non-English text before moderation.
- D
Integrate the Content Safety API with Azure Logic Apps for automated workflows.
- E
Enable geolocation filtering to analyze where the content originates from.
Show answer and explanation
Correct answers: A, B
Explanation
To implement a text moderation solution with Azure AI Content Safety, you need to enable text content detection for offensive language and hate speech. Additionally, configuring severity levels allows you to classify violations and determine appropriate actions, such as blocking or flagging content. These features directly address the requirements of the scenario, whereas the other options are either unrelated or not necessary for the described solution.
- A. Correct.
This is correct because enabling text content detection for offensive language and hate speech is a core functionality of Azure AI Content Safety and directly addresses the requirement.
- B. Correct.
This is correct because configuring severity levels for violations allows you to classify and handle content appropriately (e.g., blocking, flagging, or allowing), meeting the given requirements.
- C. Incorrect.
This is incorrect because while translating non-English text might be useful in some scenarios, it is not explicitly required for detecting offensive language or hate speech in this case.
- D. Incorrect.
This is incorrect because integrating with Azure Logic Apps is useful for automation but is not directly related to the core requirement of detecting and classifying violations.
- E. Incorrect.
This is incorrect because geolocation filtering is not a feature of Azure AI Content Safety and does not address the specific text moderation requirements.