AI-900 Describe features of Natural Language Processing workloads on Azure — Timed 10 Questions

Question 1

A language teacher uses Azure AI Language to automatically analyze hundreds of student essays. The teacher wants to identify the main topics discussed in each essay and also understand the overall sentiment (positive, negative, or neutral) expressed. Which two prebuilt Azure AI Language features should the teacher use together to accomplish this goal?

10 minute time limit — choose an answer to begin.

Accepted Answer

Key phrase extraction and Sentiment analysis. The teacher needs two features: one to identify main topics and one to analyze sentiment. Key phrase extraction extracts key phrases that represent the main topics, and sentiment analysis provides the overall sentiment. Option A is the only correct combination. Option D includes entity recognition, which identifies named entities like people or places, not the main topics discussed. Options B and C do not correctly pair a topic-identifying feature with sentiment analysis.

Answer

Entity recognition and Language detection

Answer

Text summarization and Key phrase extraction

Answer

Sentiment analysis and Entity recognition

Question 2

A hotel chain wants to analyze thousands of guest reviews to understand the overall tone of feedback (positive or negative) and to extract the most commonly mentioned features (e.g., 'room cleanliness', 'staff friendliness', 'breakfast'). Which two Azure AI Language features should they combine?

Accepted Answer

Sentiment analysis and key phrase extraction. The scenario requires two objectives: understanding the overall tone (positive/negative) and extracting commonly mentioned features. Sentiment analysis directly provides the tone. Key phrase extraction identifies important phrases like 'room cleanliness'. Therefore, the correct combination is sentiment analysis and key phrase extraction (Option A). Option C (key phrase extraction and entity linking) lacks sentiment analysis, so it does not fully address the requirement. Entity linking helps disambiguate entities but does not capture tone, making Option C incomplete.

Answer

Entity recognition and language detection

Answer

Key phrase extraction and entity linking

Answer

Text summarization and sentiment analysis

Question 3

A customer support team receives emails in multiple languages. They want to automatically determine the language of each email and then extract key phrases to summarize the issue. Which two Azure AI Language features should they use in sequence?

Accepted Answer

Language detection and key phrase extraction. The scenario requires first identifying the language of each email (using Language Detection) to enable accurate processing of multilingual content. Then, to summarize the issue, Key Phrase Extraction directly extracts key phrases from the text. The stem explicitly asks for 'extract key phrases to summarize the issue,' so only Language Detection followed by Key Phrase Extraction (Option C) achieves this. Option B uses Entity Extraction, which identifies entities like names or dates but does not extract key phrases for summarization. Options A and D are incorrect because Sentiment Analysis is not used for summarization, and starting with Entity Extraction without language detection is less effective.

Answer

Sentiment analysis and key phrase extraction

Answer

Language detection and entity extraction

Answer

Entity extraction and sentiment analysis

Question 4

A customer service team wants to analyze chat transcripts to understand customer sentiment and identify the most frequently discussed topics. Which two Azure AI Language features should they combine to achieve this?

Accepted Answer

Sentiment analysis and key phrase extraction. To understand customer sentiment and identify frequently discussed topics, the required features are sentiment analysis (to detect emotional tone) and key phrase extraction (to surface important terms/topics). Option A provides this pair. Option B (language detection and entity extraction) does not include sentiment analysis, so it cannot assess customer sentiment. Options C and D are also unsuitable because they lack sentiment analysis and do not directly address both requirements.

Answer

Language detection and entity extraction

Answer

Text summarization and question answering

Answer

Named entity recognition and translation

Question 5

A customer support team wants to automatically analyze incoming emails to (1) determine the overall emotional tone (e.g., frustrated, satisfied) and (2) identify specific key phrases that indicate the reason for contact (e.g., 'return item', 'refund policy'). Which two Azure AI Language features should they use? (Choose two.)

Accepted Answer

Sentiment analysis. Sentiment analysis is correct because it evaluates text to determine the overall emotional tone, such as frustration or satisfaction, by assigning a sentiment score (positive, negative, neutral, or mixed) at the document and sentence level. Key phrase extraction is correct because it identifies the most relevant phrases in the text, such as 'return item' or 'refund policy', which directly map to the customer's reason for contact. Entity recognition identifies predefined entities like people or locations, and language detection determines the language, neither of which addresses the requirements.

Answer

Entity recognition

Answer

Language detection

Question 6

What is 'semantic role labelling' (SRL) and why is it important for language understanding?

Accepted Answer

Identifying who did what to whom, when, and where by assigning roles to sentence participants. Semantic role labeling (SRL) is a natural language processing technique that identifies the predicate-argument structure in a sentence, assigning roles such as agent, patient, instrument, location, and time to sentence constituents. It is important for language understanding because it enables a machine to answer 'who did what to whom, when, and where,' which is foundational for tasks like information extraction, question answering, and event detection. Note that Azure AI services do not expose SRL as a primary API capability; it is a more specialized NLP technique.

Answer

Labelling training data with semantic tags for custom NLP model training

Answer

Classifying the semantic category (positive, negative, neutral) of each sentence role

Answer

Assigning roles and responsibilities to team members in an AI project

Question 7

What is 'abstractive question answering' vs 'extractive question answering' in Azure AI Language?

Accepted Answer

Extractive returns a quoted text span; abstractive generates a new synthesised answer. Extractive question answering (QA) identifies and returns a verbatim text span from the source document as the answer, while abstractive QA generates a new, synthesized answer in natural language that may not appear verbatim in the source. Azure AI Language's custom question answering is extractive; abstractive QA is typically performed by generative AI models.

Answer

Extractive works on short texts; abstractive handles long documents

Answer

Abstractive QA is faster because it doesn't need to search the full document

Answer

Extractive is pre-built; abstractive always requires custom model training

Question 8

What is 'Azure AI Bot Service' and how does it relate to Azure AI Language?

Accepted Answer

The conversation infrastructure platform that uses Azure AI Language for NLP understanding. Azure AI Bot Service is a comprehensive platform for building, deploying, and managing conversational bots. It relies on Azure AI Language (specifically the Language Understanding (LUIS) service) to perform natural language processing (NLP) — interpreting user intents and extracting entities from utterances. This integration allows the bot to understand and respond to natural language input, making option B correct.

Answer

A hosting service for training NLP models on conversational data

Answer

A monitoring service that tracks bot conversation quality metrics

Answer

Azure's billing service for tracking bot API call usage and costs

Question 9

A healthcare organization needs to extract specific data elements (such as patient names, medication dosages, and dates) from unstructured doctors' notes. Which Azure Cognitive Service is best suited for this task?

Accepted Answer

Text Analytics. Text Analytics (now part of Azure AI Language) is the correct service because it provides pre-built entity extraction capabilities specifically designed to identify and extract named entities like people (patient names), quantities (medication dosages), and dates from unstructured text. This aligns directly with the requirement to extract specific data elements from doctors' notes without needing custom model training.

Answer

Language Understanding (LUIS)

Answer

Translator Text

Answer

Speech

Question 10

A hospital wants to create a system that can transcribe doctor-patient conversations in real time and also extract medical conditions, medications, and dosages from the transcribed text. Which combination of Azure AI services should they use?

Accepted Answer

Speech to Text and Text Analytics for Health. The scenario requires real-time transcription of doctor-patient conversations, which is handled by Azure Speech to Text, and then extraction of medical entities like conditions, medications, and dosages from the transcribed text, which is specifically provided by Azure Text Analytics for Health. Text Analytics for Health is a specialized container or API within Azure Cognitive Services that uses medical ontologies (e.g., UMLS, SNOMED CT) to extract clinical entities, unlike the standard Text Analytics API which only extracts general entities like names or locations.

Answer

Speech to Text and Text Analytics API (standard)

Answer

Translator Text and Language Understanding (LUIS)

Answer

Speaker Recognition and Question Answering