[MICROSOFT AI FUNDAMENTALS (AI-900) EXAM] – EXAM-STYLE QUESTIONS
AND ANSWERS | VERIFIED AND WELL DETAILED ANSWERS | PLUS RATIONALES
| GUARANTEED PASS | 2026/27 LATEST UPDATE | EXAM PREP | STUDY GUIDE |
PRACTICE TEST
SECTION ONE: QUESTIONS 1-50
1. A retail company is building a chatbot to handle customer inquiries. The
chatbot must be able to understand the intent behind various customer
messages, such as "Where is my order?" and "I haven't received my package."
Which core Azure AI service should be used to interpret the user's goal from
the text?
A. Azure Cognitive Search
B. Azure Translator
C. Azure Bot Service
D. Azure Language Understanding (LUIS)
Correct Answer: D. Azure Language Understanding (LUIS)
Rationale: LUIS is a language understanding service that allows applications to
interpret user intent from text or speech. In this scenario, it is the appropriate
service to map the user's phrases to a specific intent, such as "CheckOrderStatus."
Azure Cognitive Search is for indexing and searching documents. Azure Translator
is a machine translation service. Azure Bot Service provides the infrastructure to
host the bot but does not perform the language understanding itself.
2. A medical research team needs to identify patterns in thousands of patient
records, including unstructured doctor's notes, to predict potential drug
,interactions. Which Azure AI service is specifically designed to extract
structured information from unstructured text like clinical documents?
A. Azure Form Recognizer
B. Azure Cognitive Services for Language - Text Analytics
C. Azure AI Document Intelligence
D. Azure Cognitive Services - Custom Vision
Correct Answer: C. Azure AI Document Intelligence
Rationale: Azure AI Document Intelligence (formerly Form Recognizer) automates
data extraction from forms and documents. It uses machine learning to understand
the layout and extract key-value pairs and tables, making it ideal for processing
semi-structured and unstructured documents like medical records. Text Analytics is
used for sentiment, key phrases, and entity recognition but is not designed for
complex form understanding. Custom Vision is for image classification and object
detection.
3. Which of the following describes the primary purpose of the Azure Machine
Learning designer when building a predictive model?
A. To write and execute custom Python or R scripts in a cloud environment.
B. To provide a drag-and-drop interface for creating, training, and deploying
machine learning pipelines.
C. To manage the versioning and storage of large datasets used in model training.
D. To automatically tune hyperparameters for any machine learning algorithm
without user input.
Correct Answer: B. To provide a drag-and-drop interface for creating, training,
and deploying machine learning pipelines.
,Rationale: The Azure Machine Learning designer is a low-code/no-code tool that
provides a visual canvas to build machine learning workflows. Users can drag and
drop datasets and modules to create a pipeline, which can be trained, evaluated,
and deployed. While it can integrate with custom scripts, its primary value is in its
visual interface for simplifying the ML lifecycle. Azure ML also offers other features
for script execution, data management, and hyperparameter tuning (HyperDrive),
but these are not the designer's primary purpose.
4. A company is using Azure Cognitive Services to analyze customer feedback.
They need to identify the prevailing emotion (positive, negative, or neutral) in
each submitted review. Which feature of the Text Analytics API is most suitable
for this task?
A. Key Phrase Extraction
B. Sentiment Analysis
C. Named Entity Recognition
D. Language Detection
Correct Answer: B. Sentiment Analysis
Rationale: Sentiment Analysis is the specific feature within the Text Analytics API
designed to assess the sentiment expressed in a text and provide a confidence score
for positive, negative, and neutral classes. Key Phrase Extraction identifies
important terms. Named Entity Recognition identifies and categorizes entities (e.g.,
people, organizations). Language Detection identifies the language of the input text.
5. A data scientist is building a classification model to predict if a financial
transaction is fraudulent. They have a dataset with a very high proportion of
legitimate transactions compared to fraudulent ones. What is the most
significant challenge presented by this dataset?
, A. Feature Collinearity
B. Data Overfitting
C. Class Imbalance
D. Concept Drift
Correct Answer: C. Class Imbalance
Rationale: A dataset where one class (legitimate transactions) vastly outnumbers
the other (fraudulent transactions) is a classic example of class imbalance. A model
trained on this data will likely be biased toward predicting the majority class and
will perform poorly at identifying the minority class (fraud). Feature collinearity is
when features are highly correlated. Overfitting occurs when a model learns the
training data too well, including its noise. Concept drift is a change in the
underlying statistical properties of the target variable over time.
6. A mobile app for a zoo uses a feature that allows visitors to take a photo of
an animal and receive information about it. Which Azure Cognitive Services
service should be used to identify the species of the animal in the image?
A. Azure Cognitive Services - Computer Vision
B. Azure Cognitive Services - Face
C. Azure Cognitive Services - Custom Vision
D. Azure Cognitive Services - Video Indexer
Correct Answer: A. Azure Cognitive Services - Computer Vision
Rationale: The Computer Vision service provides pre-built models for image
analysis, including object detection. While it may not have a specific pre-built
model for every animal species, the "object detection" feature is the correct pre-built
service to use. Custom Vision is used for creating custom image classifiers when
AND ANSWERS | VERIFIED AND WELL DETAILED ANSWERS | PLUS RATIONALES
| GUARANTEED PASS | 2026/27 LATEST UPDATE | EXAM PREP | STUDY GUIDE |
PRACTICE TEST
SECTION ONE: QUESTIONS 1-50
1. A retail company is building a chatbot to handle customer inquiries. The
chatbot must be able to understand the intent behind various customer
messages, such as "Where is my order?" and "I haven't received my package."
Which core Azure AI service should be used to interpret the user's goal from
the text?
A. Azure Cognitive Search
B. Azure Translator
C. Azure Bot Service
D. Azure Language Understanding (LUIS)
Correct Answer: D. Azure Language Understanding (LUIS)
Rationale: LUIS is a language understanding service that allows applications to
interpret user intent from text or speech. In this scenario, it is the appropriate
service to map the user's phrases to a specific intent, such as "CheckOrderStatus."
Azure Cognitive Search is for indexing and searching documents. Azure Translator
is a machine translation service. Azure Bot Service provides the infrastructure to
host the bot but does not perform the language understanding itself.
2. A medical research team needs to identify patterns in thousands of patient
records, including unstructured doctor's notes, to predict potential drug
,interactions. Which Azure AI service is specifically designed to extract
structured information from unstructured text like clinical documents?
A. Azure Form Recognizer
B. Azure Cognitive Services for Language - Text Analytics
C. Azure AI Document Intelligence
D. Azure Cognitive Services - Custom Vision
Correct Answer: C. Azure AI Document Intelligence
Rationale: Azure AI Document Intelligence (formerly Form Recognizer) automates
data extraction from forms and documents. It uses machine learning to understand
the layout and extract key-value pairs and tables, making it ideal for processing
semi-structured and unstructured documents like medical records. Text Analytics is
used for sentiment, key phrases, and entity recognition but is not designed for
complex form understanding. Custom Vision is for image classification and object
detection.
3. Which of the following describes the primary purpose of the Azure Machine
Learning designer when building a predictive model?
A. To write and execute custom Python or R scripts in a cloud environment.
B. To provide a drag-and-drop interface for creating, training, and deploying
machine learning pipelines.
C. To manage the versioning and storage of large datasets used in model training.
D. To automatically tune hyperparameters for any machine learning algorithm
without user input.
Correct Answer: B. To provide a drag-and-drop interface for creating, training,
and deploying machine learning pipelines.
,Rationale: The Azure Machine Learning designer is a low-code/no-code tool that
provides a visual canvas to build machine learning workflows. Users can drag and
drop datasets and modules to create a pipeline, which can be trained, evaluated,
and deployed. While it can integrate with custom scripts, its primary value is in its
visual interface for simplifying the ML lifecycle. Azure ML also offers other features
for script execution, data management, and hyperparameter tuning (HyperDrive),
but these are not the designer's primary purpose.
4. A company is using Azure Cognitive Services to analyze customer feedback.
They need to identify the prevailing emotion (positive, negative, or neutral) in
each submitted review. Which feature of the Text Analytics API is most suitable
for this task?
A. Key Phrase Extraction
B. Sentiment Analysis
C. Named Entity Recognition
D. Language Detection
Correct Answer: B. Sentiment Analysis
Rationale: Sentiment Analysis is the specific feature within the Text Analytics API
designed to assess the sentiment expressed in a text and provide a confidence score
for positive, negative, and neutral classes. Key Phrase Extraction identifies
important terms. Named Entity Recognition identifies and categorizes entities (e.g.,
people, organizations). Language Detection identifies the language of the input text.
5. A data scientist is building a classification model to predict if a financial
transaction is fraudulent. They have a dataset with a very high proportion of
legitimate transactions compared to fraudulent ones. What is the most
significant challenge presented by this dataset?
, A. Feature Collinearity
B. Data Overfitting
C. Class Imbalance
D. Concept Drift
Correct Answer: C. Class Imbalance
Rationale: A dataset where one class (legitimate transactions) vastly outnumbers
the other (fraudulent transactions) is a classic example of class imbalance. A model
trained on this data will likely be biased toward predicting the majority class and
will perform poorly at identifying the minority class (fraud). Feature collinearity is
when features are highly correlated. Overfitting occurs when a model learns the
training data too well, including its noise. Concept drift is a change in the
underlying statistical properties of the target variable over time.
6. A mobile app for a zoo uses a feature that allows visitors to take a photo of
an animal and receive information about it. Which Azure Cognitive Services
service should be used to identify the species of the animal in the image?
A. Azure Cognitive Services - Computer Vision
B. Azure Cognitive Services - Face
C. Azure Cognitive Services - Custom Vision
D. Azure Cognitive Services - Video Indexer
Correct Answer: A. Azure Cognitive Services - Computer Vision
Rationale: The Computer Vision service provides pre-built models for image
analysis, including object detection. While it may not have a specific pre-built
model for every animal species, the "object detection" feature is the correct pre-built
service to use. Custom Vision is used for creating custom image classifiers when
