AI-900 Practice Question: Describe features of Natural Language Processing workloads on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Automatically renaming Azure resources with appropriate names

  Why it's wrong here

  Custom NER is a text extraction tool that labels spans of text with entity categories; it has no capability to manage or rename Azure resources. Renaming resources like virtual machines or databases is a resource management operation performed through Azure CLI, PowerShell, or Azure Policy, not through an NLP model. While a custom NER model could theoretically extract resource names from documentation, it would never execute a rename action, so this option describes an infrastructure automation scenario, not entity recognition.

• ✓

  Training models to recognize domain-specific entity types unique to your business

  Why this is correct

  Custom NER in Azure AI Language lets you define your own entity categories—such as product codes, medical diagnostics, or legal reference IDs—and train a model using labeled examples of your domain text. Once trained, the model can be deployed as an endpoint that extracts these custom entity spans from unstructured documents. This is the correct answer because it precisely describes the feature's purpose: building a domain-specific entity recognizer tailored to your business vocabulary, rather than relying solely on prebuilt generic entities.

• ✗

  Replacing personally identifiable information with pseudonyms

  Why it's wrong here

  Replacing personally identifiable information (PII) with pseudonyms is an anonymization or masking operation that transforms the text, whereas custom NER only identifies and labels entity spans without altering the original content. Azure AI Language's prebuilt PII detection feature can flag PII for redaction, but the actual pseudonym replacement would require additional code or a separate service like Azure Cognitive Search's skillset. Custom NER extracts defined entity types from your own training data; it never replaces or modifies the text it processes.

• ✗

  Detecting when text contains company-specific named brands

  Why it's wrong here

  Custom NER can indeed be trained to recognize company-specific brand names as one type of custom entity, but that is only a single, narrow use case. The core function is to define and extract any domain-specific entity categories that matter to your business—such as product codes, legal clauses, or clinical terms—not just brands. Moreover, the phrase 'detecting when text contains' suggests a binary classification task, whereas custom NER performs token-span extraction, identifying exactly where each entity appears and which trained category it belongs to.

Go deeper

Related to this question